NON-EDITED VERSION SUPPORT TO JOBS FOR YOUTH IN NIGER A RETROSPECTIVE EVALUATION OF JOBS PROJECTS 2007-2018 1 NON-EDITED VERSION Acknowledgements. This report was requested by Cellule d’Analyse des Politiques Publiques et d’Evaluation de l’Action Gouvernementale of the Republic of Niger, under the supervision of Abdoulahi Garba Modibo. It was prepared by a team comprising Boureima Gado, Jan von der Goltz (TTL), Mira Saidi and Abdoulaye Sambo Soumaila. Joelle Dehasse (Country Manager, Niger), Alvaro Gonzalez (Principal Economist, Jobs Group), Christophe Lemière (Program Leader Human Development), and Ian Walker (Manager, Jobs Group) provided guidance. The team gratefully acknowledges the support of CAPEG and many Government counterparts and international partners who met with the team to provide guidance and shared information. The team also thanks the peer reviewers Thomas Bossuroy (Senior Economist, Social Protection and Jobs), Diletta Doretti (Senior Operations Officer, Finance, Competitiveness, and Innovation), and Jose Lopez-Calix (Program Leader Equitable Growth, Finance, and Institutions) for their helpful feedback, and is thankful for comments and advice provided by Mahamane Maliki Amadou, Amadou Ba, Stephanie Brunelin, Johanne Buba, Stanislas Honkuy, Emilie Jourdan, Alexandre Laure, and Ali Sidi. This report has been made possible through a grant from the World Bank’s Jobs Umbrella Trust Fund, which is supported by the Department for International Development/UK AID, and the governments of Norway, Germany, Austria, the Austrian Development Agency, and Swedish International Development Cooperation Agency Sida. 2 NON-EDITED VERSION Contents EXECUTIVE SUMMARY ...................................................................................................................................... 7 1- BACKGROUND AND SCOPE OF THE STUDY ................................................................................................. 11 2- METHODS ............................................................................................................................................. 13 3- LIMITATIONS OF THE STUDY APPROACH AND AVAILABLE INFORMATION ..................................................... 15 4- THE JOBS SITUATION FOR YOUNG NIGERIENS: A PORTFOLIO OF LOW -PRODUCTIVITY ACTIVITIES, LIMITED ACCESS TO HOUSEHOLD BUSINESS ACTIVITIES , AND ASPIRATIONS FOR A WORKING LIFE OUTSIDE OF AGRICULTURE 16 5- AN OVERVIEW OF JOBS SUPPORT IN NIGER .............................................................................................. 19 6- HOW MUCH HAS BEEN INVESTED IN JOBS PROJECTS, AND HOW MANY NIGERIENS HAVE RECEIVED JOBS SUPPORT ? ..................................................................................................................................................... 22 6.1 Investments in jobs support ....................................................................................................... 22 6.2 Beneficiaries reached ........................................................................................................................ 25 7- WHAT DIFFERENCE HAS JOBS SUPPORT MADE FOR BENEFICIARIES? ............................................................ 29 7.1 New jobs or new job activities .......................................................................................................... 30 7.2 Increased revenue............................................................................................................................. 32 7.3 Training ....................................................................................................................................... 34 8- DO PROJECTS IN NIGER ADDRESS THE MAIN JOBS CONSTRAINTS ? ARE THEY IN LINE WITH BEST PRACTICES ? . 36 8.1 Portfolio composition and cost of jobs support actions............................................................. 36 8.2 Rural development and value chains ......................................................................................... 39 8.3 Training projects ......................................................................................................................... 41 8.4 Capital-based support: cash grants, LIPW, and cash transfers................................................... 42 9- HOW HAS THE INSTITUTIONAL ENVIRONMENT WORKED FOR JOBS SUPPORT ? ............................................. 44 9.1 Many government entities are involved in overseeing jobs policies and projects, with the potential to pose implementation problems.......................................................................................... 44 9.2 The National Employment Promotion Agency strives to support employment in diverse ways, but is constrained to serving small slivers of the labor force ................................................................. 46 9.3 Even though the Government of Niger has been supporting jobs for a long time, there is still room for improvement in lessons learnt and the direction of the project portfolio ............................. 46 10- SUMMARY AND POLICY RECOMMENDATIONS FOR JOBS SUPPORT .......................................................... 48 REFERENCES .................................................................................................................................................. 53 3 NON-EDITED VERSION LIST OF FIGURES Figure 1 Self-reported obstacles to household business activities .......................................................... 18 Figure 2 Youth Aspirations and Economic Activity by level of education ................................................ 19 Figure 3 Distribution of finalized and ongoing projects by region ........................................................... 20 Figure 4 Distribution of projects by objectives ......................................................................................... 21 Figure 5 Distribution of projects by type of job support .......................................................................... 21 Figure 6 Total investment of projects per year (USD) ............................................................................... 23 Figure 7 Total and average budget by partner .......................................................................................... 24 Figure 8 Distribution of Projects by Budget .............................................................................................. 25 Figure 9 Estimated number of beneficiaries (allowing for double-counting) and scale comparisons .... 27 Figure 10 Estimation of the number of all beneficiaries (primary or secondary) among youth and comparison to the number of entrants to the labor market .................................................................... 28 Figure 11 Estimation of the number of primary beneficiaries among youth, and comparison to the number of entrants to the labor market ................................................................................................... 28 Figure 12 Distribution of beneficiaries reached by type of benefit .......................................................... 30 Figure 13 Temporary and permanent jobs created and in pipeline ......................................................... 31 Figure 14 Achieved and potential temporary and permanent jobs for youth ......................................... 32 Figure 15 Reported yield and revenue change over the course of the project, with evaluation approach .................................................................................................................................................................... 33 Figure 16 Distribution of trainees by type of training............................................................................... 35 Figure 17 Budget share by project type of non-youth projects ................................................................ 38 Figure 18 Budget share by type of project for youth specific projects..................................................... 38 Figure 19 Budget per beneficiary comparison by type of project and beneficiaries ............................... 39 Figure 20 Distribution of projects by Ministry .......................................................................................... 44 Figure 21 Distribution of youth specific projects by Ministry................................................................... 44 Figure 22 Timeline of the Government’s flagship technical training programs ....................................... 47 LIST OF TABLES Table 1List of stakeholders the team met with......................................................................................... 13 Table 2 Population distribution by age group ........................................................................................... 16 Table 3 Stylized project types .................................................................................................................... 37 Table 4 Characteristics of agricultural household and access to community infrastructure .................. 40 Annex Table 5 Reported short-term increase in revenue among project beneficiaries.......................... 57 4 NON-EDITED VERSION DEFINITIONS Youth: following the definition of youth of the Government of Niger, this study considers ‘youth’ to refer to the age range from 15 to 35. However, some of the projects included in the review consider beneficiaries aged up to 40 years of age as youths. It is also worth noting that in practice, the concept of ‘youth’ in Niger is less associated with a definite age range than with a sense of not yet being fully settled, or having assumed the full responsibilities and privileges associated with adulthood. Job: a job is work for wage or salary, in cash or in kind, as well as unpaid work for profit or family gain, whether as an employee or self-employed. Temporary job: a temporary job refers to a cash for work opportunity or a labor intensive public works (LIPW) opportunity for the purposes of this report. These tend to be short in duration, with most lasting between six weeks and six months of work. When projects provide LIPW as a primary activity, they intend them to prevent households from falling further into poverty in periods of adversity such as droughts. Permanent job: the concept of a permanent job in this report does not necessarily refer to a single primary job that by itself is enough to provide a livelihood, much less to wage employment. Rather, it mostly refers to new self-employed income generating activities that have the potential to be sustained for a longer time after support ends. Unemployment: the ILO definition considers an individual to be unemployed if they are out of work but are actively looking for a job, and available for work. Household enterprise: for the purposes of this report, a household enterprise is a durable business activity that is informal in nature, does not use hired workers, and relies on unpaid family work. Primary beneficiary: those who directly benefit from individual job support through a project. In the case where households are the main beneficiaries, we assume one primary beneficiary per household. Secondary beneficiary: beneficiaries that benefit from a project as a member of a targeted household (but are not the primary beneficiary), or as members of a community that benefits from access to new infrastructure. 5 NON-EDITED VERSION LIST OF ACRONYMS AFdB Banque Africaine de Développement AFP Agence Française de Développement/ French Development Agency ANPE Agence Nationale pour la Promotion de L’Emploi CAPEG Cellule d’Analyse des Politiques Publiques et d’Evaluation de l’Action Gouvernementale FAFPA Fond d’Appui à la Formation Professionnelle FAFPCA Fond d’Appui à la Formation Professionnelle Continue et l’Apprentissage IFAD International Fund for Agricultural Development MSMEs Micro and Small Enterprises ODA Official Development Aid OECD Organization for Economic Cooperation and Development ONEF Observatoire Nationale de l’Emploi et de la Formation Professionnelle PAFPCA Programme d’Appui à la Formation Professionnelle Continue et l’Apprentissage TVET Technical Vocational Education and Training WFP World Food Program 6 NON-EDITED VERSION EXECUTIVE SUMMARY Scope of the study and background The purpose of this study is to review the effectiveness of projects that supported jobs for youth in Niger over the years 2007-2018 . This report reviews the design and performance of 50 jobs projects carried out under the aegis of the Government of Niger over the years 2007-2018. These projects support individuals or communities; the review does not cover macro-level policies. The review has sought to assess (1) the level of investment and the number of beneficiaries reached; (2) the success of jobs programs according to existing evaluations; and (3) whether jobs programs have targeted the primary constraints to jobs for young Nigeriens and been designed in accordance with international evidence. For most young Nigeriens, a job means a portfolio of low-productivity activities, mostly in agriculture, with few opportunities to expand into more rewarding tasks or find the type of non-farm employment youth aspire to. Three key observations from the World Bank’s 2017 Niger Jobs Assessment, and recent studies of the private sector (IADD, 2017; INS, 2015) provide the context for our analysis and recommendations. Firstly, (1) for the 500,000 young Nigeriens who enter the labor market each year, a ‘job’ is nearly always a portfolio of part-time or seasonal low-productivity activities, most often in agriculture, which is the primary sector of employment for 78 percent of young Nigeriens. Secondly, (2) most households have a small casual business activity to generate some cash income – what one could call a ‘household enterprise’. However, youth rarely run these business activities, including because it is difficult to accumulate enough own savings to do so. Finally, (3) 90 percent of young Nigeriens aspire to more productive jobs outside of agriculture – but this is at odds with the reality of where work opportunities currently are. How much has been invested, and how many Nigeriens have received jobs support? Through significant investment, jobs projects have reached an estimated 11m beneficiaries over the past ten years; however, due to strong population dynamics, only about one in every thirteen young labor market entrants receives individual assistance. The projects reviewed have invested or are planning to invest about $1.6bn in jobs support through rural development and value chain projects, livelihoods support, training, and entrepreneurship projects. Yearly investment has recently risen to about $170m; this is a significant level of spending and is equivalent to about 14% of all aid to Niger. In consequence, jobs projects have reached an important number of beneficiaries – 11 million Nigeriens, including an estimated 2.9m youth. Among these, some 2.6m (including 0.8m youth) have received targeted individual support. Back of the envelope comparisons suggest that, with some assumptions, one in every six Nigerien workers may have benefited from individual jobs-related assistance. However, because many young Nigeriens enter the work force every year and because much job support is targeted toward (older) household heads, individual support reaches a smaller fraction of entrants into the labor force – about one in every thirteen. What difference has jobs support made for beneficiaries? Many projects report that beneficiaries saw an increase in revenues, but data quality is problematic, and projects must assess much more carefully whether they have succeeded in making a lasting change in jobs outcomes. Most beneficiaries (83% of the total) received support that was intended to raise the revenue generated by their activities – mostly in agriculture. Such projects typically report 7 NON-EDITED VERSION through before-and-after comparisons that the revenue of their participants increased by about half over the course of the project (equivalent to an estimated $246 per year in a poor household). This is a large increment in absolute and relative terms, but data quality is questionable, and results must be viewed with caution. Further, there is virtually no information on whether results are sustained. Given persistently low aggregate productivity in agriculture in Niger, this is a crucial question for projects to assess. With generally large households, projects should also ask how revenue increases for the family benefit youth. In past projects, support to job creation has mostly focused on temporary work opportunities offered as part of humanitarian relief, but newer projects focus increasingly on permanent jobs and on supporting youth. Job creation has in the past been largely limited to temporary work opportunities in labor-intensive public works (LIPW) and has to date created opportunities for an estimated 35,000 beneficiaries per year. These temporary jobs have mostly been made available as part of humanitarian relief after drought. They provide significant if temporary income support to poor households: a typical project pays $120-360 in total wages per participant, compared to a baseline mean annual household income of about $460 in one recent project. Newer projects seem to have shifted their focus from LIPW toward promoting jobs that could be more permanent through cash grants or in-kind inputs. They are also much more likely to target youth. Nearly 180,000 Nigeriens, most of whom are youth, have received jobs support through training, but project monitoring is inconclusive as to whether this support helped improve job outcomes. Training projects have supported about 180,000 Nigeriens, among whom three quarters were youth. Training is mostly delivered outside of formal systems, and is thoughtfully oriented toward the realities of job opportunities in Niger. However, projects provide next to no information on whether training has improved any jobs outcomes – for instance, whether trainees were able to raise their revenues, or whether they were able to work in the fields they trained. Since training programs have by far the highest expenditure per beneficiary ($2,500 at the median, compared to $1,600 for capital-based support), this is a very significant gap in monitoring. Do projects in Niger address the main jobs constraints, and are they in line with best practices? The large investment in rural development is well targeted toward the key constraints to job opportunities, but there may be more opportunities to include youth than have been seized. Jobs projects that work with Nigeriens of all ages focus heavily on rural development (80% of the total budget). The spectrum of projects ranges from those that seek to raise production of staple subsistence crops to those that look to build markets around cash crops and support value chains. With low productivity in staple crops, it remains important to help households make investments in basic proven technologies. In addition, however, projects should seek more opportunities to invest in the types of activities in agricultural value chains that are most likely to engage youth, given the low interest of this group in working in subsistence agriculture. Technical training as a mainstay of youth jobs support needs additional scrutiny, and capital-based support may better target constraints. Those jobs projects that are dedicated specifically to supporting young Nigeriens favor training (65% of the total budget) over other kinds of support. Training projects speak to the low skill levels among young Nigeriens, and most are thoughtfully designed to respond to the reality of what kind of work young beneficiaries might find. However, international evidence on the effectiveness of training projects is not encouraging, and since training is costly, projects must 8 NON-EDITED VERSION investigate much more carefully whether their support allows young beneficiaries to resolve the constraints they face. It is worth considering whether capital-based assistance would in some instances be a more effective way of supporting youth jobs outcomes. The evidence is significantly more positive, and access to cheap and risk-free capital is clearly a constraint on young workers’ ability to start modest business activities. How has the institutional environment worked for jobs support? The definition of jobs policy and the oversight of jobs projects is made more challenging by the involvement of no fewer than 14 ministries and ongoing changes in the institutional landscape. Despite a concentration of project oversight in the Ministries of Agriculture (42% of projects) and Technical and Professional Training (19%), no fewer than eleven different government bodies are involved in directing the projects reviewed, and at least fourteen ministries contribute to the definition of jobs policy. To add to this complex institutional environment, four ministries are directly in charge of the youth employment question, with no clear co-ordination mechanism. This poses challenges to staffing, strategic oversight and learning. Further, the institutional landscape has changed repeatedly over the past decade, with attendant risks for institutional memory and confusion over mandates. The National Employment Promotion Agency strives to support employment in diverse ways, but is constrained to serving small slivers of the labor force. The National Employment Promotion Agency (ANPE) provides a job matching service in the formal sector that some 27,000 Nigeriens sought to use in 2016. However, it works with a small and select group of job seekers that is much more educated than most youth, and provides placements almost exclusively oriented toward the social services (75%) and extractives sector (23%). Its current budget would not permit it to extend its services to large constituencies. In addition, the ANPE has initiated a wide range of additional active labor market policies, including some that are potentially useful to a broader and less elite group of job seekers. Yet, its most innovative initiatives to train out-of-school youth on the job in MSMEs, and support self- employment through mentoring and small grants have been limited to working with very few beneficiaries, due to lack of funding and expertise, as well as design issues. Policy recommendations for jobs support (1) Managing the jobs support portfolio: 1. Seek additional funding commitments to sustain the current level of jobs investment in the coming years. 2. Identify a coordination mechanism for the multiple ministries involved in overseeing jobs support for youth. 3. Consolidate mainstream jobs support into larger projects, and orient smaller projects to trying out innovative types of support. 4. Study whether additional jobs support should be allocated to the regions of Diffa and Tillabéry, and whether projects can better target parts of the country affected by instability. 9 NON-EDITED VERSION (2) Offering young Nigeriens job support that is effective and speaks to their aspirations and the constraints they face: 5. Rural development projects should redouble efforts to combine productivity investments with support to jobs in food system value chains, in order to engage youth unenthusiastic about work in subsistence agriculture and improve market access for households. 6. Youth jobs projects would benefit from shifting some resources from pure training to capital- based support to address the obstacles youth face in setting up market-lined activities. 7. Projects should consider setting more ambitious targets for youth inclusion, but recognize that with limited resources, higher youth participation may mean less support to household heads. (3) Learning from experience in jobs support: 8. To make jobs support more effective, projects must do better in collecting information on impacts, sustainability, and cost efficiency of jobs support; this can be done without imposing undue cost. In particular, long-standing rural development projects should revisit former beneficiaries and study how they have fared since they last received support. 9. Training programs must subject their impacts and cost-efficiency to much more serious scrutiny, given the high cost per beneficiary and mixed record of training in promoting jobs outcomes. 10. Projects should more explicitly address the gender gap in jobs and incomes and reflect these considerations in project design and outcomes. 11. Projects should systematically track how many women and youth receive support, and what their outcomes are. 10 NON-EDITED VERSION 1- BACKGROUND AND SCOPE OF THE STUDY 1.1 Background Jobs are crucial to development, poverty reduction, and stability in Niger. While Niger has recently recorded years of solid growth, it remains among the world’s poorest countries, with the sixth-lowest GDP per capita (PPP) in 2017, and a poverty rate of 45% at $1.90 per day and 82% at $3.20 in 2014 (WDI). With little structural transformation, 84% of the people continue to live in rural areas, and manufacturing contributes a mere 6% to value added. Progress in providing better economic opportunities is all the more urgent given that Niger has one of the highest population growth rates in the world, and hence, one of the youngest populations. Nearly three in five Nigeriens are below the age of 18, and no fewer than 500,000 young Nigeriens enter the labor market each year (World Bank, 2017). Providing good-enough job opportunities for these large cohorts of young women and men will be decisive for development and continued social stability in a context where idle and disenfranchised youth are vulnerable to criminalization and radicalization. This is a particularly important concern for Niger given the growing level of violence and conflict along border areas with several neighboring countries, which create an additional urgency to respond to the aspirations of the country’s youth. . More productive jobs are needed to provide perspectives for a growing young population with high aspirations. Few Nigeriens are outright unemployed, but many experience underemployment (62%), and work temporary or short-term jobs (70%) (World Bank, 2017). With most jobs low in productivity, most households maintain a ‘portfolio’ of activities to make a livelihood. Options for young people are particularly constrained, due to limited access to family farm land, hiring in the informal sector based on family relation or other relationships, the virtual absence of a formal sector, and high barriers to saving the necessary capital to establish other productive activities. At the same time, young Nigeriens have high aspirations for their economic lives, and are looking for more productive activities than the ones currently available to them. Supporting jobs for youth in Niger means addressing difficult questions of the right level of ambition, of reaching the necessary scale, and of trade-offs with funding for poverty reduction. Jobs policy for youth in Niger faces a difficult balancing act. It must be grounded in the reality that given the very modest starting point, economic opportunities for most Nigeriens are likely to remain limited in the medium term. At the same time, jobs policy needs to contend with the fact that young Nigeriens have aspirations for a different economic life than that lived by their elders. The challenge is thus to determine what kind of support is realistically feasible, and can provide a good-enough perspective for young Nigeriens. Secondly, policy must ask whether the actions envisaged can make a real difference at scale, given the very big needs for better jobs. Finally, in an environment where poverty remains pervasive, it is important to acknowledge that providing jobs support specifically for youth may mean forgoing other investments that may be more efficient in reducing poverty. To make informed choices on these questions, policy makers may need to understand clearly the effectiveness of different jobs support modalities, their appropriateness in addressing Niger’s challenges, and their cost efficiency. This study seeks to provide such an assessment. 1.2 Scope of the study This study was undertaken at the request of the policy analysis unit under the Prime Minister’s Office of the Republic of Niger. The review report was requested by the Cellule d’Analyse des Politiques Publiques et d’Evaluation de l’Action Gouvernementale (CAPEG) at the Prime Minister’s Office of the 11 NON-EDITED VERSION Republic of Niger. Its scope was designed to help inform jobs policy, and its methods were validated with the CAPEG’s scientific advisory council. The purpose of the study is to review retrospectively the effectiveness and efficiency of jobs projects carried out under the aegis of the Government of Niger over the years 2007-2018. This study answers three key questions: (1) What has been the level of investment and the number and characteristics of beneficiaries reached with different types of programs, and have program activities been carried out as planned? (2) In existing evaluations, have jobs programs been successful? (3) Have jobs programs targeted issues identified as primary constraints to jobs for young Nigeriens? Has their design been aligned with international evidence on successful programs? It reviews past and current support to jobs through projects launched between 2007 and 2018. The scope of the review extends only to projects carried out under the aegis of the Government – that is, it excludes, for example, projects implemented directly by NGOs. Fifty finalized and ongoing projects were reviewed. Of these, 26 projects provided data on results, outcomes and to a significantly lesser extent impact. The review is limited to projects that give support to individual beneficiaries or communities, and does not consider the macro-fiscal environment for jobs and the role of major infrastructure. In the interest of maintaining a manageable review scope and permit the use of a single methodology, this study is limited to jobs support at the ‘micro’ and ‘meso’ level – that is, to activities that assist individual beneficiaries or communities. It does not consider the role in supporting jobs of macro-fiscal policy, the business environment, or investments in infrastructure. This is in no way to suggest that these areas do not offer real and important opportunities to promote jobs in Niger. In particular, while Niger has improved its standing in the Doing Business index, and scores slightly better than the regional average for sub-Saharan Africa, infrastructure deficits remain pronounced (World Bank, 2019). Access to electricity stands at 16% (WDI, compared to 49% on average in IDA countries), 13% of communities have access to improved roads, and mobile subscriptions stand at 41%, (as compared to 74% on average in IDA countries). In reviewing the impact of jobs support, this study considers job creation, improvements in productivity, and access to jobs for youth. The World Bank’s Jobs Group has developed a conceptual framework1 on how to think of and evaluate jobs outcomes in operations. It maps key jobs outcomes to three categories: job creation, job quality, and access to jobs for disadvantaged groups. These are the categories we use in this study to assess jobs impacts. We discuss below how each type of outcome should be thought of in the context of jobs for youth in Niger. In brief, we argue that job creation chiefly means new self-employed activities; that quality is crucially about greater productivity and higher and more reliable income from self-employed activities; and that for our purpose, we think mainly of access for youth to activities that are more productive. An additional category called ‘intermediate outcomes’ is also used for indicators which ‘are facilitating factors that contribute to achieving the final outcomes towards more, better, and inclusive jobs.’ In the context of Niger, these relate chiefly to skills, market access, and access to finance. These categories seek clarity and a realistic view of what jobs mean to 1 World Bank (2017). Monitoring and Evaluation for Jobs Operations. Jobs M&E Toolkit, Volume 1. 12 NON-EDITED VERSION workers, and how we can support jobs, rather than focusing narrowly on the concept of job creation in a context where it may not be meaningful to do so. 2- METHODS The portfolio review was undertaken in four phases, from identification of potentially relevant projects to prioritization, data collection and conversion, and data analysis. To ensure a thorough portfolio review, we (1) identified through online research and numerous meetings with counterparts all potentially relevant projects implemented in Niger during the study period and under the Government’s aegis, by the World Bank, and other bilateral and multilateral partners. Following this identification process, we (2) undertook a prioritization exercise to select the most relevant projects for our analysis. The third phase consisted of (3) a document collection exercise to obtain the necessary information, and conversion of descriptive information from project documents into a database. Finally, (4) we undertook a thorough analysis of all available data based on indicators and categories we developed. Between February and May 2018, we met with government, bilateral and multilateral partners to identify and collect relevant project data. To identify all relevant projects, we relied on online and public sources, but also mapped all relevant stakeholders, and, starting in February 2018, undertook a series of meetings with them. The goals were (1) to introduce the study to counterparts; (2) to determine and validate what relevant projects existed; (3) to collect all relevant project documentation; and (4) to gain an understanding from clients and project implementers about the challenges in supporting jobs. Visits continued until October 2018 and included several follow-up discussions where necessary to consult and ensure that all relevant and available documentation was collected Table 1 presents the list of stakeholders we met with during this period. Table 1 List of stakeholders the team met with Government counterparts and Nigerien partners International partners Ministère de L’Agriculture et de l’Elevage GIZ Ministère de l’Emploi du Travail et de la Coopération Suisse Protection Social ANPE Lux Dev ONEF French Development Agency Ministère de la Formation Professionnelle et IFAD Technique Ministère du Commerce (Directeur PME) Agence de Développement Belge Maison de l’Entreprise Swisscontact Ministère du Plan et de l’Aménagement European Union 3N UNDP CIPMEN 13 NON-EDITED VERSION We identified more than 80 potentially relevant projects, and prioritized among them to identify the most pertinent were ones. Alongside the partners, we identified more than 80 projects which directly or indirectly impact jobs and youths in Niger. We sorted them into three categories of priority, and kept those in the first category for the portfolio review, while those in the second category were reviewed for context where the opportunity arose. The categories were defined as follows: (1) Priority 1 (projects must satisfy all criteria) ▪ Project directly tackling jobs (not necessarily youth employment), or one of its main components or activities is linked to jobs (creations, productivity, training etc.); ▪ The project has a sizable budget; ▪ It is implemented by the Government or under meaningfully direct government oversight; and ▪ Sufficient documentation is available. (2) Priority 2 (not in category 3, and any of the following apply) ▪ Not directly tackling jobs but has some small employment focused component; ▪ Highly pertinent projects which are very small in size (budget or duration); and ▪ Implemented by the Government for the most part. (3) Priority 3 (any of the following apply) ▪ Jobs component too small or very indirect (legislative or policy-oriented projects, for example); and ▪ Not under the aegis of the Government. Fifty projects were selected for document search and further analysis. Among the projects identified, 50 fulfilled the Priority 1 criteria and were analyzed further. The team undertook a significant effort to collect all relevant and available project documentation from partners. In particular, we sought to obtain (1) project concept notes and documents for the purposes of better understanding the objectives, design, and budgets of the projects, and (2) all available implementation progress and evaluation reports, including implementation status reports, final reports, monitoring and evaluation reports, impact evaluations. The documents used are listed in the references section of this report. •Potentially 80 pertinent projects •Pertinent 50 projects under Priority 1 •Projects with 26 evaluations and results 14 NON-EDITED VERSION Relevant project information, from project design to jobs results, was converted into quantitative indicators and collected in a database. The database sought to capture all descriptive project information which would provide information relevant to the three research questions. This included project scale (in terms of budget and beneficiaries), project scope (objectives, desired results), project type (components, sub-components, activities), geography (villages and communities targeted), and project implementation (timeline, challenges, delays). In addition, detailed information regarding the indicators monitored and results achieved was captured whenever possible, and classified into different categories of results. The database includes targets for performance, achieved results, and qualitative information from final reports and evaluations. 3- LIMITATIONS OF THE STUDY APPROACH AND AVAILABLE INFORMATION While we strived to identify all relevant projects in our study period, we found that institutional memory is relatively limited. This study sought to undertake a retrospective analysis of projects for the period 2007 to 2017. While we identified many projects that operated between about 2012 and 2017, projects implemented prior to 2012 were much harder to come by. Even during interviews with counterparts, there was a tendency to focus on more recent projects, and the impression of limited institutional memory of older ones. Therefore, while we do not have any specific indication that we missed early projects, it is possible that this may be the case. There were significant gaps in the available documentation, and evaluations were particularly hard to come by. We continued the search for relevant documents through the duration of the study, including through ongoing engagement with partners. The many documents amassed through this process provided rich information. However, there were also significant gaps in the information available, to the degree that some projects had to be excluded for lack of documentation. Four key gaps were identified: • There was little information on how the implementation of projects unfolded, on challenges projects may have encountered, and any changes they may have made along the way. • Budgets were available for nearly all projects. However, the available breakdown of spending rarely allowed us to identify how much of the budget was allocated to specific project components, which limited our ability to undertake a more thorough analysis of the level of investment in and cost efficiency of particular types of support. • Most importantly, only 26 projects had documentation on results, whether in the form of monitoring reports linked to the results frameworks or of impact evaluations. Because of this, we were not able to study results achieved on some kinds of support, notably those related to entrepreneurship projects. In addition (as we discuss further below), some of the estimates we present of the overall support provided for jobs in Niger are likely to be conservative. • Finally, jobs indicators are not disaggregated by gender systematically. While most projects have a specific target for reaching women, and clearly seek to monitor the number of women included, there is little information on what jobs support women beneficiaries received. 15 NON-EDITED VERSION 4- THE JOBS SITUATION FOR YOUNG NIGERIENS: A PORTFOLIO OF LOW -PRODUCTIVITY ACTIVITIES, LIMITED ACCESS TO HOUSEHOLD BUSINESS ACTIVITIES , AND ASPIRATIONS FOR A WORKING LIFE OUTSIDE OF AGRICULTURE Three key features of the jobs situation of young Nigeriens inform our analysis. Our discussion of jobs support is informed by the jobs situation for young Nigeriens, obstacles they face in finding more productive jobs, and their aspirations. In this section, we summarize three key elements that bear most directly upon our recommendations. We draw chiefly upon the World Bank’s 2017 Niger Jobs Assessment, in addition to recent studies of the private sector (IADD, 2017; INS, 2015); unless otherwise indicated, the Jobs Assessment is the source for statistics and figures shown in this section. 1. For young Nigeriens in a labor market with many competitors and few full-time productive options, a ‘job’ is nearly always a portfolio of low-productivity activities, most often in agriculture. Niger has to provide opportunities for 500,000 young workers who enter the labor force every year. Young people between 15 and 34 years of age make up 30 percent of Niger’s population (UN population data). An additional 50 percent of the population are below 15 years old, the largest share of all comparator African countries, driven by the world’s highest birth rate. Due to these powerful population dynamics, more than 500,000 young workers enter the labor force every year. The majority of young workers aged 15 to 24 have no schooling, although thanks to progress in the education system, they are much more likely to have been in school than older generations (40 percent, compared to seven percent of 55-64 year-olds) (World Bank, 2017). Table 2 Population distribution by age group Age group Population share Under 15 50% 15-34 30% 35-64 17% 65 and above 3% Source: UNFPA, 2018 For most young Nigeriens, a ‘job’ means a portfolio of income-generating activities. Jobs in Niger are virtually always informal, with a wage employment share of 4% - of which, 76% in the public sector. They are also often seasonal or temporary (70%), as well as part-time (62%). In this setting, it is more meaningful to think of a job as an individual’s and a household’s portfolio of income-generating activities (Blattman and Ralston, 2016). Working in different activities can help manage risk, but more fundamentally, it is a way of adjusting to changes in the productivity of different activities, and piecing together a livelihood from different sources that individually would not suffice. For instance, a 2014 survey showed that nearly one in three workers previously active in manufacturing shifted to agriculture during harvest time, reflecting the casual nature and low productivity of many manufacturing activities. Conversely, eleven percent of men (and very few women) reported having temporarily migrated to cities for work in the idle season. 16 NON-EDITED VERSION Work in agriculture is the norm for young Nigeriens, complemented by casual self-employed activities; most available activities are not very productive. Fully 91% of all households in Niger have at least one member who is active in agriculture. Agriculture is the primary activity for 81% of workers of any age, and a secondary source of income for an additional five percent. Young Nigeriens are no different: 78% work primarily in agriculture. Those not active in agriculture are most likely to be working in casual self- employed activities (15% of all workers). As argued above, transitions between these activities are fluid. Many jobs in Niger are not sufficiently productive to allow for a satisfactory standard of living: the overall poverty rate at $3.10 per day is at 82%. Conditions in agriculture are particularly difficult. Two out of five Nigeriens working in the sector have no cash income. Even when we leave aside those who work without pay, and only consider workers who do have some earnings, the self-employed in the agriculture sector earn three times less than those who work in other sectors. Yet, even among those who work in informal business, rather than in subsistence agriculture, 60% live in poverty. 2. Most households have a ‘household enterprise’, but youth rarely run these business activities, likely because it is difficult to accumulate enough own savings to do so Most households have a ‘household enterprise’, often in small-scale services, that can contribute meaningfully to their revenue. Most households in Niger have a business activity beyond agricultural production – a ‘household enterprise’. A household enterprise is informal in nature and refers to people working as self-employed in the non-agricultural sector with no hired workers, and utilizing unpaid family members (World Bank, 2011). Most of these activities are in small-scale services: some 37 percent are in retail trade, followed by other services (21 percent), agro-processing (16 percent), hospitality (6 percent), and personal care services, such as shoe repairs (6 percent). While the majority (58%) of household business activities operate year-round, it is the exception for them to be the primary occupation (15%). Still, they can be important to the household’s welfare and prospects by adding a complement to subsistence agriculture: expenditure of households with diverse jobs portfolios was about 25% higher than expenditures of pure farming households in 2014 (a correlation, not a causal effect). Youth rarely run household enterprises, in part because it is difficult to accumulate enough own savings to do so. It is most often household heads (56%) or their spouses (31%) who own the household’s main non-farm business activity. It is much rarer for their children to be the owners (9%). By the same token, young people under 25 years of age owned 9% of all household enterprises in 2014, but constituted 30% of the labor force. One key factor that makes it hard for young Nigeriens to set up a business activity is the difficulty in supporting the initial cost: individuals overwhelmingly (81%) rely on their own savings in financing their business activities. Presents and informal loans make up most of the balance (16%); hardly any (1%) use formal credit. High interest rates imply serious risks and make it unsurprising that few businesses use credit – the 2015 Enquête Nationale showed that the implied interest rate in small (average of $200) informal loans was 26%. The hurdles to youth in setting up household enterprises are a serious obstacle to their prospects, given how few other productive job opportunities there are in the economy, especially when considering the small share of wage or formal employment. Households face many difficulties in their business activities, but limited market demand stands out as the most common, while lack of skills is rarely mentioned. Households experience a number of constraints in their business activities, but more than two thirds among them identify limited demand for their products and services as their principal obstacle (38% cite a lack of clients, and 30% cite competition – two sides of the same coin). Indeed, most household enterprises are looking for 17 NON-EDITED VERSION customers in their local markets (73 percent) and buy any inputs locally (72%). The next most frequently mentioned challenges are the poor availability inputs, equipment, and places of business (25%), another facet of weak product and services markets, followed by limited access to finance (21%). Skills are rarely mentioned, and productivity analysis shows no correlation between the revenues generated by household enterprises and formal education, though it also demonstrates that numeracy is associated with 30 percent higher revenues. Figure 1 Self-reported obstacles to household business activities 80 70 60 50 40 30 20 10 0 Lack of clients, Lack of inputs, Weak access to Poor technical Other too much equipment, place finance and business competition of business skills Source: World Bank, 2017 MSMEs operate on a similarly limited scale as household enterprises, and face similar constraints. A survey of 700 informal household businesses of one or two employees in Niamey (IADD, 2017), as well as Niger’s 2012 Employment and Informal Sector Survey shows that small businesses nationwide and household enterprises in the capital operate at small scale and face similar constraints. Thus, about 75% if household enterprises in Niamey are active in commerce and services (18% in manufacturing); very few (5%) have invested to expand over the five preceding years, and the overwhelming majority finance themselves through own or family savings (97%). Recruitment of workers is mostly through personal relationships (73%) – a finding that reflects how difficult it is for young Nigeriens to access jobs outside of their own family’s activities in the urban informal business sector. 3. Most young Nigeriens aspire to the more productive jobs outside of agriculture – but this is at odds with the reality of where there are work opportunities The jobs aspirations of young Nigeriens are at odds with the reality of employment structures. More than 90 percent of youth aspire to work outside of the agricultural sector; they have particularly little interest in working in subsistence agriculture (as opposed to more intensive and market linked agriculture). At least half of youth with any schooling wish for a job in the formal sector. The overarching goal of changing their jobs outlook also translates into plans for change in the medium term. Thus, most young Nigeriens report that over a time frame of five years, they would like to find a different job, or resume their studies; far fewer aspire to continue their current work or grow their activity. At the same time, the reality for nearly all young Nigeriens is work in agriculture or in casual self-employment, and 18 NON-EDITED VERSION the odds against a transition out of farm work remain steep for Nigeriens whose fathers were farmers – one to twelve for women, and one to nine for men. Figure 2 Youth Aspirations and Economic Activity by level of education 100% 80% 60% 40% 20% 0% Aspired Actual Aspired Actual Aspired Actual No edu. Primary edu. Secondary edu. Agriculture Self-employment Formal private Formal public Source: World Bank, 2017 5- AN OVERVIEW OF JOBS SUPPORT IN NIGER The following analysis is based on a review of available documentation for 50 jobs-related projects in Niger over the past decade. Not all of the projects included have the explicit objective of tackling jobs. However, given what a job in the Niger context entails, the projects have been selected because they each seek to support jobs outcomes in a meaningful way. Similarly, only 15 of the projects discussed in the following sections have youth as their primary target group; yet, none are designed in a way that obviously excludes young beneficiaries, and a significant number set specific targets for the inclusion of young beneficiaries, or report on results they have delivered for Number of projects: 50 young beneficiaries. Nearly half of the projects included in the review have been completed (45%), with the balance ongoing Projects dedicated to youth: 15 (39%) or planned (16%). Projects on average are active for four Projects which have a youth specific years, though in many instances, the projects in the portfolio target: 4 are continuations of each other which seek to either deepen Average duration: 4 years or expand support and coverage. For example, IFAD’s three flagship projects, PUSADER, Ruwanmu and ProDaf are Completed projects: 45% expansions of one another, as are the World Bank’s Community Action Programs phase 2 and 3. Thus, some of the projects included in the review reflect long-standing jobs support efforts. The projects in the portfolio provide jobs support in all regions of Niger, though there is no clear focus of investment in the stressed regions of Diffa and Tillabéry. A reasonable number of projects have been carried out and are active in each of the eight regions of Niger. It is notable, however, that among 19 NON-EDITED VERSION ongoing and planned projects, there are somewhat fewer projects in the regions of Tillabéry and Diffa, despite the needs in these regions due to displacement and conflict spillover (Figure 3 – please note that unless otherwise noted, the source for all tables and figures is World Bank staff analysis of the project portfolio database compiled for this report). It is worth recalling that this is the regional pattern among jobs projects only; there might be significantly more development projects targeting these regions, for instance, focused on other needs such as humanitarian assistance and peace building. There are also fewer projects in Niamey, likely due to the focus of many projects on jobs in agriculture. Regrettably, an analysis of spending by region is not possible with the available documentation; it could help shape a more insightful picture of the real investments made. Figure 3 Distribution of finalized and ongoing projects by region 30 25 20 14 15 14 10 10 8 13 9 10 10 5 14 10 12 11 11 7 7 5 0 Acheve en cours/non-debute Source: World Bank staff analysis of the project portfolio database compiled for this report Most jobs-relevant projects in Niger frame their overall goals as reducing poverty and food insecurity, rather than looking at jobs outcomes as an end in themselves. According to their stated development objectives, projects most commonly target food security (35%), an improvement in livelihoods (25%), or poverty reduction (17%). Indeed, few projects directly target ‘job creation’ (12%), or other outcomes related to productivity as their main goal. These objectives are in line with the reality of widespread poverty and smallholder agriculture, as well as overall government priorities which seek to create more self-sustainability against hunger and resilience to drought. Still, it is worth noting that actions to support jobs are largely framed as a matter of improving household welfare, rather than of improving the economic prospects of youth. 20 NON-EDITED VERSION Figure 4 Distribution of projects by objectives 35% Improviding living conditions 25% 17% Economic growth 15% 13% Professional and technical training 13% 12% Capacity building 10% 10% Local development 6% 6% Climate change 4% 0% 5% 10% 15% 20% 25% 30% 35% 40% Among the different ways to improve jobs, most projects focus on raising revenues or improving human capital. Based on an analysis of project descriptions of the goals they set themselves in supporting jobs, few projects (8%) explicitly attempt to create jobs. It is much more common for projects to either aim to improve job quality by raising revenues (78%), or to create better opportunities for jobs by improving skills (54%) or market access (34%). This finding is perhaps less surprising than it may at first appear: in Niger, the difference between ‘creating a job’ and ‘raising revenues’ is fluid. For instance, projects that support and fund the creation of a new part-time income generating activity in which households transform an agricultural product and sell it in the market would most likely argue that they have increased revenue, or that they have improved market access. Yet, to the degree that we think of ‘job creation’ as ‘more productive tasks for workers to do that they could not previously undertake’, such projects might also plausibly argue that they have changed the set of available activities, and hence in a sense, created a new job within the household. Conversely, as we discuss below, projects may claim to have ‘created a job’ when they have provided a temporary work opportunity that may be better thought of as ‘assistance in raising revenues.’ It is, however, notable that skills (‘human capital’) are by far the most important intermediate outcome considered. Figure 5 Distribution of projects by type of job support 21 NON-EDITED VERSION 6- HOW MUCH HAS BEEN INVESTED IN JOBS PROJECTS, AND HOW MANY NIGERIENS HAVE RECEIVED JOBS SUPPORT ? Given widespread poverty and the desire of youth for better opportunities, it is important to ask whether jobs projects provided support at a scale that is commensurate with needs. Even as youth in Niger express a desire for different economic opportunities than those their parents faced, poverty and food insecurity in Niger remain widespread. The gulf between the reality of youths’ economic lives and their aspirations thus remains very wide. To make a difference to the jobs outlook for young Nigeriens, both support to productivity as well as support to qualitative changes in jobs activities need to operate at large scale. In the following section, we ask whether the projects operated at a scale that is largely commensurate with needs. For context, it is important to recall that the jobs support projects studied here are only one dimension of jobs support; macro-fiscal policy and policy on economic fundamentals also play a decisive role in achieving scale of jobs support, but is not part of our focus. 6.1 Investments in jobs support Projects included in the review invest a total of $1.6bn in jobs support over the years 2007-2024, including $318m in projects solely dedicated to youth. The estimate of overall investment is likely conservative, because of missing budget data for some projects, because we may not have captured all relevant projects in the early years, and because additional projects will be added to be pipeline by 2024. Slightly more than half of the total investment (53%) has come in the form of grants, while the balance is (usually highly) concessional loans. Completed projects represent 26% of this budget (US$ 425 million), though this figure clearly underrepresents what has already been spent given that a breakdown of spending for ongoing projects is limited. Projects that are exclusively dedicated to supporting youth make up 20% of the overall investment, though youth will undoubtedly also be among the beneficiaries of other projects. At least $170m per year has been spent over the past three years, an increase over earlier investments, and equivalent to 14% of overall ODA to Niger each year. We assess spending trends by assuming that projects spend an equal share of their budget every year while they are active (this of course not accurate – projects typically ramp up spending, and to a lesser degree, phase it out). We are less confident on spending in the early years of the study period, and believe that some of the apparent early increase in spending shown in Figure 6 may be an artefact that is due to missing data. However, it seems clear that jobs spending has increased significantly in the 2010s, following the 2010 and 2012 droughts. Over the past three years since 2016, we estimate spending to have been at least $170m per year. For comparison, overall official development aid (ODA) to Niger in 2017 was USD 1.2bn, so that jobs-related investments represented some 14% of the total (OECD). Projected spending from the year 2020 onward shows that further jobs projects would have to be added to the pipeline in order to maintain the current level of investment. 22 NON-EDITED VERSION Figure 6 Total investment of projects per year (USD) 200,000,000 180,000,000 160,000,000 140,000,000 120,000,000 Potentially incomplete information Pipeline not yet fully 100,000,000 defined 80,000,000 60,000,000 40,000,000 20,000,000 - 2012 2024 2005 2006 2007 2008 2009 2010 2011 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 No fewer than sixteen partners are working on providing jobs support in Niger under the Government’s aegis, with the World Bank, IFAD, and the European Union accounting for most investment among the projects in our sample. Based on the sample of projects we collected, the largest share (72%) of investment comes from three organizations: the World Bank ($600m), IFAD ($300m), and the European Union ($265m). These partners, together with AFD and Coopération Suisse, also account for 70% of projects that currently provide jobs support. While some partners favor large-scale projects, the typical jobs project is small, with a median budget of $19 million, raising questions over whether the portfolio could benefit from consolidation. AfDB, IFAD, WFP, and the World Bank run relatively large jobs-related projects, with a median project budget of about $40m. However, the remainder of the projects reviewed are relatively small, despite the fact that the smallest projects (with a budget below $1m) have been excluded from our review. Thus, the median budget is US$19 million, and a third of them (36%) do not exceed US$ 15 million. Smaller projects may allow for greater experimentation. However, based on a review of project descriptions and design, the impression is that many of these projects are not unusually innovative in their design. Based on an in-depth review of activities in a sample of projects, traditional agricultural activities take precedence as most projects work on improving the productivity of rainfed agriculture, improving irrigation systems, providing higher quality seeds and fertilizers. The cultivated products are also very traditional (onions, sorghum, millet etc.). This raises serious questions of fragmentation and efficient scale for supervision and learning. 23 NON-EDITED VERSION Figure 7 Total and average budget by partner 700,000,000 70,000,000 600,000,000 60,000,000 500,000,000 50,000,000 400,000,000 40,000,000 300,000,000 30,000,000 200,000,000 20,000,000 100,000,000 10,000,000 - - sum average 24 NON-EDITED VERSION Figure 8 Distribution of Projects by Budget Jobs project budgets 15 10 5 0 0 50 100 150 200 Budget (million $) 6.2 Beneficiaries reached The projects reviewed have to date reached nearly 11 million beneficiaries in Niger. As of the time of writing, projects report having supported 11 million beneficiaries, including 2.6 million primary beneficiaries2 who directly benefit from some form of jobs support given to them individually. The remaining 8.4 million are secondary beneficiaries who benefit either as members of the households of a primary beneficiary, or as members of communities3 that have access to newly built infrastructure such as roads, irrigation systems, storage facilities, and hence, hopefully, increased revenues etc. The number of beneficiaries assisted to date represents 74% of the 3.5m primary beneficiaries, and 91% of the 12m total beneficiaries the projects have set themselves the goal of reaching by 2024. At the same time, fewer than half of the projects analyzed have been completed. It therefore stands to reason that many projects can expect to deliver support to more beneficiaries than initially assumed. Because data on beneficiaries are missing in 13 out of 45 completed and ongoing projects, the true number of beneficiaries reached is likely higher.4 It is important to note that it is difficult for us to tell whether households and individuals participated in multiple projects, and hence, to avoid double-counting. The number of total beneficiaries can therefore not be interpreted as ‘the number of Nigeriens who have 2 Assuming that one person from each targeted household is a primary beneficiary in projects that report their beneficiary numbers in terms of households rather than individuals. 3 Beneficiary data is missing for 17 projects. 4 Projects without beneficiary data account for 12% of the budget of all completed and ongoing projects. If average cost per beneficiary were the same among these projects as among those with information, we could infer that they might have reached an additional 1.4m beneficiaries; if we assume that project cost is the same within each type of project, the imputed number of additional beneficiaries is 1.6m. 25 NON-EDITED VERSION ever benefited from jobs support’. However, one way to look at potential repeat targeting is to consider that those who participated in more than one project received more meaningful jobs support. Jobs projects set themselves the goal of including women as about one in every three beneficiaries, and seem to exceed this target somewhat. Surprisingly, a significant number of small projects do not set explicit goals for the inclusion of women. However, all large projects do: they project that among a total of 13.2 million expected beneficiaries, 4.4 million (33%) should be women. In practice, projects appear to perform slightly better, and report that women account for some 39% of the 11m beneficiaries already reached. As discussed above, there is very little gender-disaggregated data on jobs support provided and associated results; this is an important information gap projects must seek to close. Projects that have set specific targets for youth inclusion intend to reach nearly one million young beneficiaries, and an additional 1.9 million young Nigeriens may benefit through other projects. The 15 projects dedicated solely to jobs for youth are small; together they aim to benefit 226,985 young Nigeriens. In addition, four larger projects set specific targets for youth inclusion, and intended to reach 744,000 young Nigeriens – 22% of those projects’ total projected beneficiaries. Youth are also likely to be among the beneficiaries of projects that do not set specific targets for youth support. Firstly, many projects that do not specify precise targets for youth inclusion nonetheless pledge in a general way to include youths. In fact, youth inclusion in jobs projects in Niger has become a targeting consideration similar to the participation of women and is often mentioned in the context of ‘paying particular focus to vulnerable groups’. Secondly, youths stand to benefit as members of households and communities. Assuming that projects that do not set explicit targets for youth inclusion have a similar share of young beneficiaries as the projects that do set a goal, an additional 1.9 million youths (22% of a total of 8.4 million remaining beneficiaries) could benefit from jobs support. We have less information available to assess how many young Nigeriens may have been the primary beneficiaries of jobs support. Certainly, all beneficiaries of projects dedicated to youth received such direct support. Assuming that in the other projects, the youth share among primary beneficiaries is the same as among all beneficiaries (22%), we would then conclude that about 750,5545 young Nigeriens have received such support. But this approach may lead to too optimistic an estimate, because it does not take into account the propensity of projects to target household heads. Back of the envelope comparisons suggest that the number of Nigeriens who received jobs support is meaningful. How meaningful are the total beneficiary numbers? For a sense of proportion, we can compare to the population, labor force, and the number of entrants. For the sake of illustration, we assume conservatively that due to double counting, there are only 5.5m individual beneficiaries among the reported total of 11m. Then, given Niger’s population of 21.5m in 2017, we would conclude that about one in four Nigeriens (26%) has directly or indirectly benefited from some kind of job support. This is clearly a significant reach of coverage; the question is how meaningful the benefits have been. With the same assumption on double counting, 1.3m primary beneficiaries reached today would be equivalent to 15% of the labor force. This is still a considerable level of coverage. 5 That is, 226,985 beneficiaries of projects dedicated to youth, and 22% of the primary beneficiaries of all other projects, i.e., 523,569 youth. 26 NON-EDITED VERSION Figure 9 Estimated number of beneficiaries (allowing for double-counting) and scale comparisons With an exceptionally high number of entrants to the labor market, it is less clear that projects have made a dent in supporting young workers For the sake of understanding the scale of youth support, it is worth recalling that given Niger’s extremely high population growth rate of 3.8%, 500,000 young Nigeriens currently enter the labor force each year (WDI). To consider how the scale of job support stacks up to this strong dynamic, suppose that an equal share of the 2.9m young beneficiaries are reached in each year over 10 years (given the increase in spending on jobs support over the past few years, this is likely to be lower than current numbers of beneficiaries). Maintaining the conservative assumption on double-counting, we would conclude that 145,000 young Nigeriens, or 29% of the 500,000 annual entrants into the labor force, have received some kind of support as primary or secondary beneficiaries each year. However, the picture is less positive if we consider the number of primary beneficiaries. With the assumptions above, only 7.5% of entrants would have received support such individual support. 27 NON-EDITED VERSION Figure 10 Estimation of the number of all beneficiaries (primary or secondary) among youth and comparison to the number of entrants to the labor market Figure 11 Estimation of the number of primary beneficiaries among youth, and comparison to the number of entrants to the labor market The relatively modest share of youth who receive individual jobs support is a consequence both of large youth cohorts and a focus on optimal poverty reduction in beneficiary targeting. The share of entrants who receive individual support is low due both to just how large the youth cohorts in Niger are, and to the limited degree to which projects specifically target youth. The number of entrants poses an immediate funding challenge: for instance, assuming a relatively modest cost per primary beneficiary of $1,000 (see Section 8), and with perfect targeting, supporting one in ten entrants would cost $50m annually, and supporting one in four entrants, $125m – about 30% and 74% of all current annual jobs support. Conversely, the degree to which projects seek to work with youth also matters: while we have noted that projects only set themselves the goal of ensuring that 22% of their beneficiaries are youth, 28 NON-EDITED VERSION the projected share of youth under 25 years of age is 34% of the labor force in 2018, and that of youth under 35 years is 59% (UNFPA 2018 and World Bank 2017). Projects may in practice include a higher share of youth than they aim for. But it is worth noting that the modest goals they set is a natural result of targeting support with the goal of maximal poverty reduction in mind. The question of what the right balance between actions to reduce poverty and to support youth cannot be answered through analysis alone, and is genuinely a political one. In the following, we seek to show how current jobs support projects addresses these important priorities, and what may be some alternative options, given the observed outcomes and lessons from international experience. In summary, significant investment in jobs support has reached many beneficiaries; however, it is important to assess the impact of such support, and to note that due to strong population dynamics, a relatively lower share of youth has received assistance. In summary, the projects reviewed have invested or are planning to invest about $1.6bn in jobs support. Yearly investment has recently risen to about $170m; this is a significant level of spending, and is equivalent to about 14% of all aid. In consequence, jobs projects have reached an important number of beneficiaries – 11 million Nigeriens, including an estimated 2.9m youth. Among these, some 2.6m (including 0.8m youth) have received individual support. Back of the envelope comparisons suggest that the level of support among Nigeriens of all ages is quite broad: with some assumptions, the data suggest that about one in every six Nigeriens may have benefited from individual jobs-related assistance. It is, however, important to ask the question is how much of a difference the support provided has made to jobs outcomes. However, because many young Nigeriens enter the work force every year and because much job support is targeted toward (older) household heads, individual support reaches a smaller fraction of young Nigeriens – about one in every thirteen. 7- WHAT DIFFERENCE HAS JOBS SUPPORT MADE FOR BENEFICIARIES ? To understand whether jobs support has made a meaningful difference for beneficiaries, we consider its reported impacts in terms of job creation, revenue increase, and the number of people trained. To assess whether projects have created real benefits, we consider three indicators that are commonly tracked by projects and can shed light on real jobs outcomes. These are: the creation of jobs, increases in revenue, and training. As pointed out above, the distinction between revenue increases and the creation of jobs or businesses is fluid in an economy like Niger, where most workers carry out baskets of income-generating activities. Results data are not available for all projects and all types of impacts and must be interpreted with caution. It is crucial to note some caveats in interpreting the reported outcomes. Firstly, only 26 out of 45 ongoing or completed projects report results. While results data are available for the largest projects in the portfolio worth $461 million, it is therefore reasonable to conclude that we likely underestimate aggregate numbers on jobs results achieved. Secondly, almost all projects only report outcomes in terms of a before-after comparison for beneficiaries. Very few compare results to a control group. Thirdly, projects nearly always report outcomes right after their activities conclude; it is therefore not possible to comment on sustainability. Finally, we also sought to assess impacts on enterprise creation and investment, but with little data available, had to abandon the attempt. (Thus, the sets of indicators that projects track are aligned with the goals they set – namely chiefly, to raise revenues and provide 29 NON-EDITED VERSION training, and to a lesser extent, to create jobs – with much less emphasis on new businesses and on investment.) By far the largest share of beneficiaries received support to raise the revenue they generate with their job, followed by training and short-term work opportunities. Of the 11 million primary and secondary beneficiaries reached so far, we have a clear understanding of what kind of job support about seven million beneficiaries have received. For the remaining 37% of beneficiaries, the available project documentation does not allow us to look at the kind of support provided in greater detail. Through those projects that we are able to analyze further, 830,000 households (equivalent to 5.8 million individuals, or 83% of beneficiaries) benefited from measures which increased their revenues by at least 30%. In addition, 94,000 households (i.e. 650,000 individuals or 9%) received monthly cash transfers to support their income. Finally, 180,000 individuals (7%) received training, 320,000 benefited from a temporary work opportunity, and 104,000 from a potentially long-term job. Figure 12 Distribution of beneficiaries reached by type of benefit 320,000 Beneficiaries 4,000,000 5,800,000 180,000 104,000 650,000 Other/unknown Increased revenues received cash transfers Temporary work opportunities Permanent work opportunities Training 7.1 New jobs or new job activities At least 276,000 temporary work opportunities and 20,000 permanent jobs have been created since 2011, with another 44,000 temporary opportunities and 84,000 permanent jobs to be created over the coming years. The available data show that projects reviewed have created or intend to create a total of about 320,000 temporary work opportunities, as well as 104,000 permanent jobs. Youth account for at least 20% (88,000) of the intended beneficiaries (mostly, under projects that have not yet concluded).6 These aggregate numbers almost certainly under-report the actual results achieved: (1) only 22 projects have the necessary evaluations or targets to allow us to assess their performance or level of ambition; (2) many projects that involve labor intensive public works do not monitor or report temporary opportunities created; (3) projects do not account for the indirect jobs created, for instance, in buying inputs for public works, or the induced jobs created through an infusion of cash into an 6 The share of youth is likely to be under-reported, since some projects include youth among their beneficiaries, but do not specifically track results for youth. The data is insufficient to impute the number of young beneficiaries of new job opportunities in the manner done above for the total number of beneficiaries. 30 NON-EDITED VERSION economy. We also note that some projects that fund income-generating activities monitor only revenues generated, and do not consider the new activities to be new jobs. Figure 13 Temporary and permanent jobs created and in pipeline 44,000 276,000 84,000 20,000 TEMPORARY PERMANENT Achieved Pipeline Since 2011, the project portfolio has provided income support through temporary work opportunities for about 35,000 beneficiaries per year on average. The projects reviewed have, since 2011, created at least 276,000 temporary work opportunities, with another 44,000 still pending. Temporary jobs are ‘cash for work’ or labor-intensive public work (LIPW). They tend to be short in duration, with most lasting between six weeks and six months of work. When projects provide LIPW as a primary activity, they intend them to prevent households from falling further into poverty in periods of adversity such as droughts. In line with this intention, LIPW typically offer 60 to 90 days of work, at wages that projects generally aim to peg to market rates for unskilled workers, say $2-4 per day. With these parameters, beneficiaries can hope to earn between $120-360 by participating in an LIPW program. Given the poor ability of good jobs, this is not an insignificant boost to revenues: the mid-point of this range is about half of the annual household income of $464 among the beneficiaries in the CAP3 project. In addition to emergency support, rural development projects provide a significant number of work-days as an ancillary benefit of financing the construction or rehabilitation of infrastructure. While temporary work opportunities must primarily be viewed as a time-bound support to incomes and consumption, there is anecdotal evidence from projects that people do reinvest the money earned into expanding or creating new income generating activities: “LIPW have made it possible to distribute (often substantial) revenues to vulnerable strata. Part of these revenues were invested in income generating activities (small commerce, fattening of small ruminants, and development of other existing activities). Women in particular have invested part of their modest revenues in economic activities.” Newer projects focus increasingly on permanent jobs and on supporting youth. In the context of the projects reviewed, the concept of a permanent job largely does not refer to a single primary job that by itself is enough to provide a livelihood, much less to wage employment. Rather, it refers to new self- employed income generating activities that have the potential to be sustained for a longer time after support ends. As we discuss below, where such activities are productive, they have the potential to meaningfully change the welfare of beneficiaries, given that a ‘job’ for most Nigeriens is a portfolio of low-productivity seasonal or part-time activities. Completed projects focused heavily on temporary employment (93%), in particular in the context of drought and other emergencies. By way of contrast, ongoing and pipeline projects project that two thirds (66%) of their beneficiaries will receive support toward permanent jobs, including the large majority of young beneficiaries (87%). Most of these jobs 31 NON-EDITED VERSION are expected to emerge from two support modalities. Firstly, funding or in-kind inputs that entrepreneurship projects provide to youth to develop their own income generating activities; indeed, half of the permanent jobs in the projects we reviewed are expected to come from a single project, the World Bank’s Youth Employment Project. Secondly, apprenticeships and (more rarely) internships that training projects provide for young beneficiaries, with the objective of transforming these into permanent jobs. Figure 14 Achieved and potential temporary and permanent jobs for youth 90,000 80,000 70,000 60,000 50,000 87% 40,000 30,000 20,000 10,000 13% - Completed Potential Temporary Permanent 7.2 Increased revenue Data gathered from six rural development and value chain projects suggests that participating farmers raised the production of targeted crops by about half. Most projects have focused either on staple crops –millet or sorghum – or traditional marketable crops such as onions and cowpeas. Very few focus on less traditional horticulture products such as tomatoes and cabbage, or internationally traded products such as peanuts or sesame. Figure 15 shows the range of outcomes for different projects and crops. As the figure also shows, it is important to note that for the most part these are before-after comparisons for beneficiaries or comparisons to small ad-hoc control groups; therefore, the real impact of the project (compared to taking no action) may be different. As is evident, there is a considerable range of changes, from a decrease of production by one-third in one case to more than doubling in two cases. Overall, participating farmers raised their production by about half (52% at the mean, 48% at the median). To achieve results, rural development projects provide several kinds of support simultaneously: training and inputs to help adopt new technologies and techniques (horticulture, irrigation), dissemination of improved seeds, provision of matching grants or facilitation of access to finance, investment in rural infrastructure (connecting roads, irrigation systems, land management) and finally, capacity building to institutions such as farmers’ associations and extension services. Given the seasonal nature of agricultural work and volatility in returns in Niger, there is a particular focus on irrigation. 32 NON-EDITED VERSION Figure 15 Reported yield and revenue change over the course of the project, with evaluation approach Source: FIDA (2013a); FIDA (2014a); FIDA (2018); World Bank (2017c); World Bank (2018c). Projects report that higher productivity went hand-in-hand with a revenue increase of at least 50% among at least 830,000 beneficiary households, but data quality is problematic and limits confidence in this assessment. Projects generally support productivity in a limited number of agricultural products. It is therefore worth asking what kind of contribution an increase in production makes to household revenue. Regrettably, data on revenues are available only from four projects. While these are some of the largest projects in the portfolio, it is therefore still likely that we underestimate the number of households that have experienced higher revenues. It is also important to recall that information on changes in revenue comes from before-after comparisons among beneficiaries, rather than from comparison with a control group. There is therefore a risk that the observed changes are due to factors other than jobs support through the projects. Based on the recorded results, beneficiary households on average increased their revenue by 53% over the course of the project (weighting projects by the number of beneficiaries, and conservatively assuming that the projects that reported an increase of “at least 30%” achieved no more than 30% - see Annex Table 5 for details). Given an average household income of $464 among the beneficiaries of one of the projects (the World Bank’s ‘CAP3’), this is an absolute income gain on the order of $246 per beneficiary household each year that benefits are sustained (for a total of $204 million per year among all beneficiary households). This is a large change in income for poor households. It is, for instance, equivalent to about twice the observed difference in revenues among farming households that use fertilizer and improved seeds and those that do not (see above). That said, with an average household size of seven one must ask, from the point of view of outcomes for youth, how much young household members stand to benefit, in particular with respect to their ability to expand their activities. 33 NON-EDITED VERSION Some households obtained supplemental income through cash transfers, notably 94,000 households who participated in a safety nets project that provided a monthly transfer of US$20. The World Bank’s Safety Net project reaches 94,000 households through cash transfers of $20 for 18 months. A well- designed impact evaluation conducted 18 months after the cessation of the transfers provides what is likely the most reliable evidence of jobs impacts available in Niger. It shows that beneficiary households increased their investment in productive assets and livestock, as well as their participation in saving groups (Mills et al., 2016). In addition, these transfers seem to have had an impact on agricultural output, with increases on the order of 30% over non-beneficiary households – akin to the effects found in rural development support.7 (The evaluation did not find an impact on household’s diversification of their portfolio of activities – perhaps not surprising, given that this is support to the poorest of the poor, who will not easily be able to start new activities). While there is not a direct measure of revenue, it is reasonable to hope that the agricultural productivity increase also raised household revenue. The issues of youth inclusion and of sustainability of revenue increases are critical and likely differ by type of projects – but there is virtually no data to assess it. Given how few households have highly productive activities, projects can meaningfully improve job quality if they succeed in supporting households to do better the things they already do, or to add new and more productive activities to their jobs portfolio. However, two observations are in order. First, given the large household sizes in Niger, it is worth asking how much young Nigeriens who have not set up their own households benefit from the revenue increases reported above. Secondly, with the exception of the Safety Net project mentioned above, none of the projects we reviewed re-interviewed beneficiaries sometime after the completion of activities to assess whether impacts persisted. Regrettably, this is a common issue in project evaluation. Yet, such an exercise would be essential to learning about project effectiveness. For instance, cash transfers inherently raise revenues temporarily. Their impact on future revenues is dependent on how these transfers are utilized and whether they are invested. Similarly, support to agricultural activities through rural development of VC projects may be effective at supporting long term revenue increases, but their sustainability depends on whether the promoted activities adequately tackle constraints. 7.3 Training Nearly 180,000 Nigeriens, most of whom are youth, have received jobs support through training. Training is focused on young beneficiaries, who make up 75% of participants. Half of the beneficiaries (50%) received training through projects whose main focus was on building technical skills. The remainder were trained as part of activities in rural development and entrepreneurship projects. Training in rural development projects either focuses on the introduction of new techniques and technologies to farmers (often through farmer field schools), or on some form of business, entrepreneurship, or soft skills curricula. Most training occurs outside of formal TVET centers. We attempted to define in more concrete terms what kind of training activities different projects provide. However, the analysis is complicated by the fact that one large project does not define what kind of training it has provided, beyond making clear that training was non-formal. 7Using reported impacts on overall agricultural output shown in Table 9 of Mills et al. (2016), compared to output among non-beneficiaries as shown in Table 2. 34 NON-EDITED VERSION 12. The largest share of trainees have participated in farmer field schools, that is, training provided in the context of an agriculture or livestock project. These courses typically seek to train farmers in the techniques to produce products supported by the project, as well as the use of inputs and technologies provided. A noteworthy innovation within this approach is the SIFA – a champs école reserved for young trainees who have been given a piece of land to farm by their household upon completion of the training. 13. Complementary trainings are intended for youth who already have some kind of training or are already working in some capacity, and wish to further hone their skill or to specialize. Examples of these trainings are sewing, mobile phone repairs, baking, motor pump maintenance, or leather goods making. 14. ‘Professionalizing’ trainings on the other hand are geared towards youth with little to no schooling or training. They are short in duration (one to three months) and seek a quick turnaround into the labor market. 15. Dual apprenticeships are much longer (two years) and the trainee undergoes both classroom and on the job learning. Generally, both the apprentice and master are provided a small stipend in these projects. Figure 16 Distribution of trainees by type of training Distribution trained Nigerien in non formal sector Distribution of trained Nigeriens SIFA 2276 Entrepreneurship 3540 Formal 15000 Dual apprenticeship 6044 Professionalizing 6059 Informal/ Complementary 14,999 non 165,202 formal Farmer field schools 33,364 Undefined 110,727 Monitoring reports are inconclusive as to whether training projects succeed in ensuring that beneficiaries are able to work productively after training ends. Based on available data from small follow up surveys, most of those who have participated in a training program are employed or self- employed; but given the virtual absence of unemployment in Niger by the narrow ILO definition of the term, this is not a very meaningful outcome. Anecdotally, one project reports that at least 70% of their trainees undertake activities related to their field of training (with no further information on the actual fields of employment). As we argue below, international evidence suggests that training programs struggle to improve jobs outcomes for their beneficiaries, and are often not cost-efficient; partners interested in providing training should very carefully consider whether it is realistic that their project can provide the kinds of results that would justify the expense. In summary, projects can point to some real results, but must assess much more carefully whether they have succeeded in lastingly changing the jobs outcomes that matter for beneficiaries. Data on 35 NON-EDITED VERSION results is only available for about half of the projects analyzed. Most beneficiaries (83%) received support that was intended to raise the revenue generated by their activities. Such projects typically report that they have raised revenue by about half (equivalent to an estimated $246 per year in a poor household). This is a large increment in absolute and relative terms, but there is virtually no information on whether results are a direct consequence of the intervention or whether they are sustained, or on how much youth benefit. Job creation has in the past been largely limited to temporary work opportunities for an estimated 35,000 beneficiaries per year. These temporary jobs provide significant income support and have mostly been made available as disaster relief. Recent projects shift their focus toward promoting jobs that could be more permanent. Finally, training projects have supported about 180,000 Nigeriens. Training is mostly informal, and it is thoughtfully oriented toward the realities of job opportunities in Niger. However, there is no information on whether training has improved any jobs outcomes; given the high cost of training programs, this is a significant knowledge gap. 8- DO PROJECTS IN N IGER ADDRESS THE MAIN JOBS CONSTRAINTS? A RE THEY IN LINE WITH BEST PRACTICES? 8.1 Portfolio composition and cost of jobs support actions Projects that support jobs in Niger can usefully be classified into five categories according to their activities: rural development, value chains, livelihoods, training, and entrepreneurship. To help understand the packages of services and activities offered by different projects, we group them into five stylized types of jobs support. These categories are (1) rural development, (2) value chains, (3) livelihoods, (4) training, and (5) entrepreneurship. Table 4 describes them. It is important to recognize that, in practice, the distinctions between categories is somewhat fluid, and therefore, the decision which category a given project should be assigned to is in some cases subjective. For instance, rural development and value chain projects lie on a spectrum of activities to raise rural productivity, with varying degrees of emphasis on subsistence farming and market access. Similarly, while there are specific ‘training’ projects, many other projects will include a subsidiary training component. 36 NON-EDITED VERSION Table 3 Stylized project types Types of projects Description Rural development • Objective: Increase agricultural income and revenues; • Support: A combination of activities and inputs including but not limited to increasing access to infrastructure, capital, training (especially non formal) and capacity building; • Target: farmers (especially household heads) Value chain • Objective: Improve competitiveness of enterprises (especially household enterprises linked to agriculture) and their interaction in the manufacturing, transport and selling for products; • Support: Access to infrastructures and capital as well as information and technical training; • Target: existing micro and small enterprises or creating new enterprises which can fill the gap in the value chain Livelihoods • Objective: Reduce poverty and food insecurity; • Support: Cash transfer or public work activities; • Target: vulnerable and or marginalized households Training • Objective: Improve beneficiary skills; • Support: Most projects offer non-formal training and emphasize dual apprenticeship, quick professionalizing training or informal agricultural training; • Target: Most of the time out of school and illiterate youths or the unemployed Entrepreneurship • Objective: Encourage new types of economic activities instead of traditional income generating ones; • Support: cash grants and entrepreneurship training; • Target: Mostly youth. About 80% of the total budget of job projects that do not specifically target youth is spent on rural development. Projects that do not specifically define youth to be their intended beneficiaries account for 80% of the total budget of all projects reviewed. Among these projects, those that support rural development make up by far the largest share of the total budget – 80%. Rural development projects also account for about 10m of the 11m beneficiaries projects have reached so far. Value chain (12%) and livelihoods (8%) projects make up the balance. Projects dedicated to jobs for youth strongly emphasize training, and invest twice as much in related projects than in capital support. Most of the projects that do not specifically target youth provide some kind of support to accessing capital, although few focus strongly on it. However, among projects that aim to work exclusively with youth, by far the largest share of investment goes toward training projects (65%). Projects that facilitate access to capital through livelihoods (an investment share of 21%) and entrepreneurship support (14%) receive only half of the funding provided to training projects. 37 NON-EDITED VERSION Figure 17 Budget share by project type of non-youth projects 12% 8% 80% Développement rural intégré Rural development Livelihoods VC Figure 18 Budget share by type of project for youth specific projects 14% 21% 65% Entrepreneurship Entreprenariat/auto-emploi Livelihoods Formation Training There are large differences in spending per beneficiary, with rural development investing by far the least, and the median training project spending more than 50 percent more per beneficiary than the median livelihoods project. Data on budgets and beneficiaries are not available for all projects. Cost per primary beneficiary is lowest by far in rural development projects (median spending of $350).8 This is not directly comparable to other project types: rural development projects generally involve investments in infrastructure and community micro-projects, and projects generally consider all households in communities that participate in the project to be among their beneficiaries. Livelihoods and training projects work with similar numbers of beneficiaries, and spending per beneficiary can be compared. Training projects incur by far the highest costs, with a median of about $2,500 per beneficiary (and at least $5,000 per beneficiary for the most expensive 25% of training projects). Livelihoods projects that provide individual support have a median budget per beneficiary of about 8 To facilitate comparison across projects, we maintain the assumption that there is one primary beneficiary per beneficiary household where necessary. 38 NON-EDITED VERSION $1,600. This discrepancy in spending per beneficiary is in line with global experience (Blattman and Ralston, 2016). Figure 19 Budget per beneficiary comparison by type of project and beneficiaries Rural development Livelihoods Training VC Entrepreneurship 0 5 10 15 Budget and number of primary beneficiaries (log) Budget per primary beneficiary Primary beneficiaries 8.2 Rural development and value chains In Niger’s overwhelmingly agriculture-based job market, jobs projects must link with agriculture to have an impact at scale – and with very low household investment and poor market access, there are obvious opportunities to raise productivity. Agriculture is the primary activity for nearly eight in ten young workers in Niger (78%). To have an impact at scale, it is therefore nearly inevitable that jobs projects support productivity in activities linked to agriculture, whether on-farm or off-farm. It is also worth noting that households and communities rarely are rarely able to invest in some proven ways to raise productivity, from cultivating cash crops to using irrigation and fertilizer to community infrastructure (see Table 4). There thus remain many clear constraints for rural development projects to tackle. Similarly, as noted, shallow demand is the most prominent constraint for household business activities, with nearly 70% citing the issue as an obstacle. Value chain support therefore has an obvious constraint to confront. 39 NON-EDITED VERSION Table 4 Characteristics of agricultural household and access to community infrastructure Characteristics of agricultural households Sells at least one crop in the market 27% Holds title to land 12% Uses irrigation 12% Uses inorganic (manure) fertilizer 18% (43%) Access to community infrastructure Permanent market 9% Periodic market 29% Bank or microfinance institution 11% Agriculture extension center 7% Cereal Bank 60% Laterite (asphalt) road 31% (13%) Source: World Bank, 2017 Rural development projects in Niger provide integrated packages of support to address multiple constraints on productivity; this makes implementation complex, but responds to clear investment shortfalls and is in line with good practice. Nearly all rural development projects reviewed provide several services, intended to alleviate multiple constraints simultaneously. They (a) teach farmers to employ new techniques and technologies; (b) provide cash or in-kind inputs such as new modified grains and pesticides; (c) build infrastructure (irrigation systems, roads linking to market, rehabilitation of lands, etc.), often by funding community micro-projects, which tend to be implemented as temporary labor intensive projects; and (d) provide some capacity building to local and national institutions. Despite the potential complexities in implementation, this approach is in line with the very low baseline level of high-productivity investments in Niger, as well as with evidence that suggests that projects that seek to tackle a single constraint in agriculture tend to have limited impacts (World Bank, 2018b). However, it is worth noting that there is little good evidence – in Niger or internationally – on what features of rural development projects help sustain benefits. Partners should seek to learn about this question. In particular, some of the long-standing projects that have gone through several permutations should revisit their early efforts, and study how their beneficiaries have fared in the years since the programs concluded. Value chain projects aim to address productivity and the key constraint of market access for household enterprises, mostly in value chains linked to agriculture. Value chain projects (a) provide modern seeds or other inputs to farmers; (b) offer funding through matching grants to enterprises and farmers; (c) work on improving quality and product diversification to help products compete in larger markets; (d) provide technical or business training; and (e) invest in infrastructure projects. While this suite of support is in some ways similar to that provided by rural development projects, VC projects focus less on raising primary productivity, and more on increasing product quality and market access. They tend may seek to support the creation of enterprises which can close the gaps in the chain – for 40 NON-EDITED VERSION instance, by giving incentives to aggregators or input merchants in taking on the fixed cost of operating in new markets. Projects that support cash crop value chains have been shown to raise revenues; while they rarely succeed in encouraging hiring by the supported farmers (given the availability of family labor), they do often succeed in stoking indirect job creation, e.g. in transport, processing and packaging, and trading (World Bank, 2018). To assure that rural development projects support youth jobs outcomes, it is worth revisiting their targeting and level of ambition. It is beyond the scope of our analysis to consider whether the details of project design (for instance, the approaches taken to extension services or to the choice of community microprojects). Rather, we note two questions partners should revisit in order to ensure that projects promote jobs outcomes for young Nigeriens. 16. Should rural development projects specifically target support to youth? Rural development projects typically view household heads as their primary clients, and rarely target youth as beneficiaries. This may be a sensible choice with a view to achieving the greatest impact on poverty, as well as a pragmatic choice, given that household heads traditionally to control the farmed land. However, if the goal is to support jobs for youth, one must ask how much the economic prospects of youth are likely to improve when household income rise. One could consider models that seek to enable youth to take responsibility for some agricultural activities. An interesting example of such an approach in Niger is the SIFA, a scheme under which young Nigeriens in rural localities receive support conditional on their household setting aside land for them to cultivate. 17. What is the right level of ambition in supporting subsistence or cash crops? As noted, the spectrum of rural development projects ranges from those that chiefly seek to raise reliable production of staple subsistence crops such as sorghum and millet to those that seek to build markets around cash crops such as horticulture products. Partners interested in supporting jobs outcomes for youth should carefully weigh whether youth may take a less dim view of work in agriculture if they are able to engage in market-linked activities, and whether they would stand a better chance of building increasingly better jobs portfolios if they had access to a cash crop. 8.3 Training projects With a low skills base, it is natural to look to technical training as a mode of jobs support to youth, but traditional training and job placement programs do not have a record of success. Education outcomes in Niger have improved greatly, but skill levels clearly must rise further. It is therefore not surprising that jobs programs have looked to promote skills – from basic literacy and numeracy, to informal training in farming skills, business management skills, and formal technical and vocational training. However, traditional training programs that focus solely on skills and pay little attention to enabling beneficiaries to apply their skills have a poor record (Card et al., 2015; Kluve et al., 2019; McKenzie, 2017b). One way to understand this evidence is to consider how well-aligned training is with the realities of the labor market young beneficiaries face: thus, even a well-trained beneficiary who may for instance have been trained in a formal TVET program, will still face low demand and steep barriers in accessing capital to set up a business activity. (None of these results, of course, diminish the importance of basic education, or suggest that training does not have an intrinsic value, beyond its impact on jobs.) 41 NON-EDITED VERSION Training programs in Niger show sensitivity to the demand for labor, but given the pervasiveness of other constraints and the high cost of training, projects should investigate their effectiveness and cost-efficiency much more closely. Many training programs in Niger seek to focus on activities their beneficiaries are likely to be realistically able to undertake – for instance, champs-écoles directed toward improving revenue from farming, and informal technical training aimed at supporting new income-generating activities. Even where training is aligned in this way with labor market opportunities, however, the international evidence suggests that partners considering training projects should ask what other constraints beneficiaries are likely to face, particularly in terms of access to capital and demand for their services. The kind of support training projects generally provide to help beneficiaries establish themselves – inserting them into internships and apprenticeships – may not directly resolve these obstacles. An obvious alternative is to test whether it is effective to combine training with grants, as is done in the World Bank’s Youth Employment project. In addition, formal training programs tend to be costly, so that even those that have proven effective may not be cost-efficient (Blattman and Ralston, 2016). Training projects should therefore very closely scrutinize their cost efficiency. 8.4 Capital-based support: cash grants, LIPW, and cash transfers Capital support – often cash – has a good track record in allowing beneficiaries to undertake more productive activities. In low-income countries, small cash grants commonly provide between $200- 1,000 in support of a beneficiary’s plans for new, more productive, income-generating activities. Programs vary in whether they provide purely cash or in-kind support (for instance, tools), and in whether they provide any other support such as training that may be complementary to cash. The evidence has been building that cash grants are effective in raising revenues, that they achieve impacts at relatively low cost, and that they avoid the problems of risk and low take-up of microfinance loans. (Blattman and Ralston, 2016; Fox and Kaul, 2017). Support to income generating business activities offers an opportunity for broad jobs support for youth, and could be a realistic way to allow youth to work at least partly on activities outside of primary production. As noted, many households conduct small business activities, but youth are very underrepresented – while youth under 25 make up nearly one-third of the labor force, they own fewer than one in ten household business activity. It is intriguing to ask whether it is not precisely small manufacturing and services activities linked to agriculture, but outside of primary production, that might be within reach for many youth with some support, and also more appealing to them than work in subsistence agriculture. One could also consider such support for small business activities for seasonal migrants in the cities of Niger. Since own savings are the chief way to finance business activities – as noted above, 97% of households use own savings or family gifts to finance jobs activities – and saving is a long and arduous process for youth in low-income households, it is sensible to consider whether cash grants could represent a good approach in helping youth establish business activities. In planning such projects, partners should consider that there is little good international evidence on (i) what it takes to make impacts sustainable; (ii) which design features can help address the problem of limited consumer demand; and (iii) whether technical or soft skills training is a useful complement to capital support, given the potentially significant additional cost. These aspects deserve careful attention, and it will be important to do more to learn from project experience. 42 NON-EDITED VERSION Entrepreneurship grants for more ambitious youth business activities could be tried at experimental scale, but may pose serious implementation challenges, and struggle to reach scale. Internationally, grants and other financing mechanisms have been used to support business activities among young people outside of agriculture and simple rural non-farm activities. This is often done through business plan competitions or similar programs. Projects provide grants or subsidized funding such as guarantees or interest-free loans. The track record is limited, but promising. However, it is important to consider the practicalities of this type of jobs support. The YouWin! program in Nigeria was highly successful – and cost-efficient – in creating jobs, but it provided grants of $50,000 on average per beneficiary (McKenzie, 2017). Accounting for the difference in GDP p.c. between Nigeria and Niger, a similarly meaningful startup support in Niger might require a grant of about $6,000 per beneficiary. Pragmatically, it is clearly a difficult task to select beneficiaries for large youth grants in a way that is fair and transparent. This is all the more true given that the preparation and evaluation of business plans requires significant capacity on the part of the applicants as well as the implementing agency. In this context, it is worth noting that the Agence Nationale pour la Promotion de L’Emploi (ANPE) has tried to launch a business plan support program, but could ultimately fund only three microprojects (as of last available 2011 data), with an average funding of about CFA 2m (about $4,000). Public works and cash transfers have been used in Niger for emergency support and to build basic resilience, consistently with the international evidence. The evidence on the jobs impact of LIPW is quite well-developed. It shows that LIPW improve consumption and other measures of current well- being (Blattman and Ralston, 2016); it also suggests that beneficiaries may be able to save some of their wages and may re-invest them in another income-generating activity (Bertrand et al., 2016), and that LIPW may help prevent dissaving in crises, which will allow beneficiaries to resume other IGA when times get better. Cash transfers can play an analogous role in helping the most marginal households build the basic level of resilience needed to consider embarking on new more productive activities. In line with this evidence, LIPW have been used appropriately in Niger for jobs support to help in overcoming or recovering from a crisis such as a drought, and to build basic resilience (as well as, in the case of LIPW, opportunistically where the construction of community infrastructure requires labor). In summary, the large investment in rural development is well targeted toward constraints and may offer additional opportunities to include youth; technical training as a mainstay of youth jobs support needs additional scrutiny. Jobs support modalities in Niger have a clear focus: projects that work with Nigeriens of all ages largely work on rural development (80% of the total budget), and projects that focus on young Nigeriens favor training (65%). Investments in rural development target important productivity constraints, though better evidence is needed of their success in helping beneficiaries overcome constraints. They should assess whether it is sensible to directly set targets for including youth, and carefully seek opportunities to support jobs in food system value chains to engage youth disaffected with farm work. Training projects are designed to respond to the reality of the demand for labor, but are costly, and must investigate much more carefully whether training allows young beneficiaries to resolve constraints. It is worth considering whether capital-based assistance would in some instances be more effective way of supporting youth jobs outcomes. 43 NON-EDITED VERSION 9- HOW HAS THE INSTITUTIONAL ENVIRONMENT WORKED FOR JOBS SUPPORT ? Government institutions have long supported jobs, but face constraints in their ability to serve youth and to guide the development of jobs support projects. 9.1 Many government entities are involved in overseeing jobs policies and projects, with the potential to pose implementation problems Many jobs projects are overseen either by the Ministry of Agriculture or the Ministry of Education. While many ministries are involved in overseeing jobs projects, due to the important role of rural development and training, the Ministries of Agriculture (42%) and Education (19%) take a particularly important role. Among projects that specifically target youth, the majority fall within the purview of the Ministry of Education, given the emphasis placed on training programs for youth. Figure 20 Distribution of projects by Ministry Ministère de l'agriculture et de l'élevage 42% Ministère des Enseignements professionnels et techniques 19% NA 12% Ministère du Plan 6% Haute Autorité à la Consolidation de la Paix 6% Ministère de l'emploi, du travail et de la sécurité sociale 4% Haut Commissariat à l'Initiative 3 N 4% Ministère de l'entreprenariat des jeunes 2% Ministère de la jeunesse et des sports 2% CNEDD 2% Cabinet du Premier Ministre 2% 0% 10% 20% 30% 40% 50% Figure 21 Distribution of youth specific projects by Ministry Ministère des Enseignements professionnels et 50% techniques Haute Autorité à la Consolidation de la Paix 19% Ministère de l'emploi, du travail et de la sécurité sociale 13% Ministère du Plan 6% Ministère de la jeunesse et des sports 6% Ministère de l'entreprenariat des jeunes 6% 0% 10% 20% 30% 40% 50% 60% 44 NON-EDITED VERSION Four ministries are directly in charge of the youth employment question, with no clear co-ordination mechanism. Even for the more narrowly defined question of youth employment, four ministries have a direct responsibility. These are (1) Ministère de l’Emploi, du Travail et de la Sécurité Sociale (DG de l’Emploi) ; (2) Ministère des Enseignements Professionnels et Techniques (DG de l’Orientation et de l’insertion Professionnelle) ; (3) Ministère de la Jeunesse et des Sports (DG de la Jeunesse et de l’Education Permanente) ; (4) Ministère de l’Entreprenariat des Jeunes. There are currently no clear coordinating mechanisms between these ministries. Defining a clear process for coordination could be a particularly helpful step toward policy consistency. No fewer than 14 ministries are involved in defining jobs policy and overseeing jobs projects, which poses challenges to staffing, strategic oversight and learning. Despite a concentration of oversight in the Ministries of Agriculture and Education, no fewer than eleven different government bodies are involved in directing the projects reviewed, and at least fourteen ministries contribute to the definition of jobs policy. These include central ministries with responsibility for setting development policy, notably the Planning Ministry. In addition to ministries mentioned above who are traditionally thought of as responsible for jobs and youth, other important line ministries are also involved: - Ministry of Trade and Private Sector Promotion ; - Ministry of Energy; - Ministry of Industry; - Ministry of Mining; - Ministry of Petroleum; - Ministry of Post Telecommunications and the Digital Economy; - Ministry of Cultural Renaissance, the Arts, and Social Modernization; - Ministry of Tourism and Crafts; - Ministry of Transport. There are clear rationales for the involvement of various government bodies in each individual project. However, when considering the overall ensemble of jobs projects, one must ask whether the multitude of actors do not stretch available human resources to thinly, and whether it could not lead to overlaps in responsibilities, and may make it difficult to collect and learn lessons, and to provide effective guidance on the strategic direction of jobs support. The institutional landscape has changed over the past decade, with attendant risks for institutional memory and confusion over mandates. Several government entities involved in overseeing youth employment support in Niger have undergone significant changes, with ministries consolidating and dividing, and new structures emerging. There is concern that the changes have weakened institutional memory as key human resources have been reassigned or let go, while there is limited physical and electronic storage of documentation on past jobs support. In addition, there is the risk of additional overlap in mandates. For instance, the Ministry of Youth Entrepreneurship was recently set up, with the mandate of promoting entrepreneurship among youths; its mandate, however, could be viewed to overlap both with the Ministry of Labor Ministy of Education, as well as their substructures, the employment agency, and the FAFPA (Fond d’Appui a la Formation Professionnelle), which seeks to implement government policy on training and insertion. 45 NON-EDITED VERSION 9.2 The National Employment Promotion Agency strives to support employment in diverse ways, but is constrained to serving small slivers of the labor force The National Employment Promotion Agency provides a job matching service but serves a small group of Nigeriens competing for formal sector jobs. Of the structures set up to support jobs for youths, the Agence Nationale pour la Promotion de L’Emploi (ANPE) under the Ministry of Labor and Social Protection is tasked with supporting and placing job seekers. Limited information is available on its performance, but some annual reports provide a sense of its impact. In 2016, 27,000 Nigeriens registered with the ANPE for job placement (compared to 11,500 in 2007), more than half of them (56%) between 18 and 29 years of age. At the same time, ANPE registered 17,000 jobs vacancies; it is unclear to what extent ANPE was able to satisfy job offers in 2016, but earlier data from 2007 showed that nearly all offers received in this year were filled. Job seekers who register with the ANPE are much more educated than the overall work force. Among them, 15% have a university degree, 29% have secondary schooling, and 24% TVET training. Similarly, most of the 2016 job offers were in the social services sector (75%) or the extractive industry sector (23%). While the agency’s budget for 2016 is not available, the 2014 budget was $1.6m. Thus, assuming that the agency filled all vacancies (and ignoring spending on other jobs services), the cost per job placement is US$95 per placement, or US$195 assuming a 50% placement – both akin to the cost per beneficiary incurred in rural development projects.9 It is clear that the agency provides a real service of job matching in the formal sector, but serves a small and select group of young job seekers. It is also clear that its current budget will not permit it to serve large constituencies. The National Employment Promotion Agency has initiated a wide range of additional active labor market policies, including some that are potentially useful to a broader group of job seekers, but they have reached very few beneficiaries. In addition to job matching services, the ANPE has developed a range of additional initiatives that could in principle be useful to young job seekers. An internship program for graduates (‘PIJD’) is geared toward a small strata of well-educated workers. It provided job experience for some 332 job seekers in 2011, and in the context of the PRODEC project, for 1,714 beneficiaries in 2014-2016. In addition, the agency has also sought to launch a range of well-considered activities that target broader groups of beneficiaries. Thus, it has sought to operate a program to train out-of-school youth on the job in MSMEs (PIPME), but as of 2011, only 35 placements had been carried out. Two other programs have set themselves the goal to support self-employment through business training, business plan development, and funding with small grants (CFA 100,000 to 3,000,000, or about $200-6,000) or partial loan guarantees. Regrettably, both programs have supported only a handful of projects as of the most recent available reports. Problems these efforts have repeatedly encountered include limited funding, a lack of experienced personnel for business plan analysis, and poor motivation among participants. 9.3 Even though the Government of Niger has been supporting jobs for a long time, there is still room for improvement in lessons learnt and the direction of the project portfolio The Government of Niger has a long history of providing jobs support through a series of flagship projects; it is crucial to ensure a productive evolution of these efforts. The Government of Niger has a long history of engaging in the various key job support modalities. In Figure 21, we show the example of Government support to technical training. The Government’s first real investment in technical training started in 1995 with the Nigetech project – a multi-year effort that segued into a second project phase in 9 The numbers are based on the 2014 budget of the ANPE which was 935,000,000 FCFA (US$1,620,000) 46 NON-EDITED VERSION 2000. In 2005, the government created the PAFPCA (Programme d’Appui à la Formation Professionnelle Continue et l’Apprentissage) to continue support provided through the Nigetech project and implement a training fund set up in 2008 (the FAFPCA – Fond d’Appui à la Formation Professionnelle Continue et l’Apprentissage). More recently, starting in 2015, the FAFPCA was transformed into the FAFPA (Fonds d'Appui à la Formation Professionnelle et à l'Apprentissage). Figure 22 Timeline of the Government’s flagship technical training programs One of the first instances which showed the commitment of the Government toward making technical training a priority Sought not only to train youths in the formal and informal sector but also put in place a national training policy that ensured the integration of disenfranchised populations Was intended to support the institutional environment for technical training through the creation of the Nigetech NGO and the creation of the first time of a Ministry dedicated to technical training. It was also intended to play an operational role to support the FAFPCA Operates like a fund for training activities and becomes a central structure of the training environment in the country Continues the funding of training activities An analysis of these flagship projects points to the importance of evaluation, iterative learning, and sustainable finance in improving the effectiveness of jobs support. The Government’s support to technical training shows a certain historical path dependency: new projects were based on the design of older ones. This is typical of several of the Government’s main jobs support modalities. It is thus worth offering some observations on the evolution of the project sequence. Our analysis is based on the available documents for the projects mentioned above, and especially an evaluation of the PAFPCA by the European Union (Bourdain and Akplogan, 2010), one of the only detailed reports available which reviews the implementation and impact of the projects. We propose the following lessons: - Document conservation and solid evaluation practice are necessary to ensure learning within series of projects. Given that high quality evaluations are nearly absent, it is difficult to understand the results and impacts of the previous projects and extract lessons in the design of successor projects. For example, while the project Nigetech 2 was subject to an evaluation, the evaluation only reports the outputs of the program. Thus, it notes that the program trained some to 30,000 artisans, workers or unemployed individuals. However, it provides no information on whether the training had any impact on the employment outcomes of the 30,000 beneficiaries, or if not, what kind of additional support would be necessary to ensure success. Such information would clearly have been highly useful in allowing subsequent efforts – starting with the PAFPCA 47 NON-EDITED VERSION – to learn from experience and improve design. At an even more basic level, significant gaps in the conservation of documents such as implementation and completion reports prevent national and international actors from having references to inform future programs. - Management challenges which impeded previous projects are likely to repeat in new projects. The scarcity of documentation makes it difficult to extract lessons learned not only regarding the design of projects, but also regarding their implementation. At the same time, if it is likely that new projects will face the same management challenges as their predecessors, unless pitfalls are identified and avoided. Thus, subsequent training projects have continued to encounter the following management issues: (1) obstacles to disbursing; (2) difficulties in forming effective partnerships with the private sector and orienting training toward market needs; (3) limited ability to align with the Ministry’s strategic priorities; and (4) poor human resources. - An absence of stable funding mechanisms limits sustainability, and follow up projects do not in themselves solve this issue. Youth employment in Niger poses challenges at a very large scale, and needs are unlikely to diminish any time soon. Hence, sustainability of support is crucial. Yet, as is the case with management issues more generally, successor projects are likely to encounter the same sustainability issues as their predecessors if there is no conscious effort to find more stable funding mechanisms. In its support to technical training, the Government made at several points an effort to devise sustainable funding mechanisms, but none has become fully effective. Thus, the NGO Nigetech was fully funded by the European Union, and while it was intended to grow into a mainstay of technical training in the country, has struggled to become self-sustainable without EU funding. Similarly, while the Government has pledged some corporate tax revenue to ensure the sustainability of the FAFPCA, funding flows to date have remained small and volatile. 10- SUMMARY AND POLICY RECOMMENDATIONS FOR JOBS SUPPORT Summary This report has reviewed the effectiveness of jobs projects in a labor market that offers young Nigeriens with few job opportunities that are productive and that satisfy their aspirations. This report has reviewed the design and performance of 50 jobs projects carried out in Niger to support individuals and communities carried over the years 2007-2018. Throughout, it has sought to acknowledge key features of the job situation that faces young Nigeriens today. Thus, while very few young Nigeriens aspire to work in agriculture, the reality is that for most of them, a job is a portfolio of activities focused on agriculture that is not very productive. At the same time, while most households have a small casual business activity to generate some cash income, youth are rarely able to run such activities. With a significant investment in jobs support, projects have provided individual support to an estimated one in six Nigerien workers – but because many young workers enter the labor market every year, providing broad support to entrants is challenging. The projects reviewed have invested or are planning to invest about $1.6bn in jobs support, with recent yearly investment reaching an estimated $170m. This investment has allowed jobs projects to give some individual support to a substantial share of the work force – with some assumptions, an estimated one in six workers. However, 48 NON-EDITED VERSION some 500,000 young workers enter the labor market every year, and jobs projects struggle to support a significant share of them individually; our data suggests that currently, about one in every thirteen entrants receives such support. It is further worth noting that with limited resources, targeting support toward youth competes with the approach of targeting support toward household heads to reduce poverty. Projects can point to some meaningful impacts, but must assess much more carefully whether they have succeeded in changing jobs outcomes that matter to beneficiaries in a lasting way. The data on results achieved by jobs projects has weaknesses, but it allows for the following conclusions: - Job creation has in the past been largely limited to temporary work opportunities that provide significant income support, mostly as part of humanitarian relief. There is a notable recent shift in funding toward projects that favor youth and promote jobs that could be more permanent. - Many projects report that they have raised the revenues of beneficiary households in the short- term by a meaningful amount. But data quality in nearly all these assessments is questionable, and results must be viewed with caution. Importantly, there is no information on whether higher revenues were sustained. - Training projects have supported about 180,000 Nigeriens, mostly in programs that thoughtfully try to equip beneficiaries with skills that are useful in the kind of work they may be able to find. However, there is no information on whether training has improved income or other jobs outcomes that matter to beneficiaries. The large investment in rural development is well targeted toward constraints but there may be more opportunities to include youth than have been seized so far; technical training as a mainstay of youth jobs support needs additional scrutiny. Jobs support through projects that work with Nigeriens of all ages focuses on rural development (80% of the total budget), and projects dedicated to young Nigeriens favor training (65%) over other types of support. Investments in rural development target important productivity constraints. However, they should assess whether it is sensible to directly set targets for including youth, and carefully seek opportunities to support jobs in food system value chains to engage youth disaffected with farm work. Training projects are designed to respond to the low skills level, but are costly and rest on weak evidence, and must investigate much more carefully whether the training provided allows young beneficiaries to overcome constraints and work in more productive activities. Capital-based assistance is backed by more encouraging evidence, and targets the steep barriers youth face in saving or borrowing funds necessary to start small market-linked activities. Small cash grants are the most obvious fit to support broad access to casual market-linked activities; entrepreneurship support programs could help better-educated youth start more ambitious activities, but will face difficulties in operating at scale. 49 NON-EDITED VERSION Recommendations (1) Managing the jobs support portfolio: 1. Seek additional funding commitments to sustain the current level of jobs investment in the coming years. The level of investment in jobs projects has increased to about $170m per year over the past three years. However, commitments for the coming years are not yet at the same level, and additional investment is needed to maintain the current scope of support. 2. Identify a coordination mechanism for the multiple ministries involved in overseeing jobs support for youth. Projects are overseen by many different ministries, with frequent changes in the institutional landscape. There is a risk that this arrangement may overstretch available human resources and make it difficult to exercise strategic oversight and ensure effective learning and good institutional memory. A credible coordinating mechanism is needed. 3. Consolidate mainstream jobs support into larger projects, and orient smaller projects to trying out innovative types of support. Many projects are quite small (one-third have a budget of less than $15m), including those that provide mainstream jobs support, rather than testing innovative approaches. It is worth trying to consolidate the portfolio to avoid unnecessary overhead and coordination cost. 4. Study whether additional jobs support should be allocated to the regions of Diffa and Tillabéry, and whether projects can better target parts of the country affected by instability. There is no clear focus of planned jobs support on stressed regions of the country, notably Diffa and Tillabéry. A high level of humanitarian aid and other support on these regions may offset the absence of such a focus in jobs investments. However, it is worth reviewing whether additional jobs support is warranted, and whether jobs projects can take the risk of instability into account in their targeting decisions. (2) Offering young Nigeriens job support that is effective and speaks to their aspirations and the constraints they face: 5. To address constrained demand and better engage youth, rural development projects should redouble efforts to combine subsistence crop productivity with investments in cash crops and value chains and to understand what activities are attractive to youth. Rural development projects make up by far the largest part (80%) of job support that is not specially targeted to youth. This is sensible: agriculture is the primary income source for most workers (81%) and will continue to be the obvious source of jobs for youth in the near term; productivity is low; and projects can report raising the crop production and incomes of their beneficiaries. However, partners should investigate whether work on cash crops or service jobs linked to food sector value chains would prove more satisfying to youth disaffected with work in subsistence agriculture. Secondly, because household enterprises overwhelmingly complain about lack of demand for their products and services (68%), partners should look for additional opportunities to complement investments in productivity with value chain support to improve market access. 6. Youth jobs projects would benefit from shifting some resources from pure training to capital- based support. Two-thirds of the budget dedicated to jobs for youth goes to training projects. However, the evidence on the effectiveness of pure training programs is not encouraging. At the same time, capital-based support could address the barriers youth face in starting income- 50 NON-EDITED VERSION generating activities in agriculture and beyond. There is a good case for shifting some resources to such support. Projects could also experiment with combinations of training with capital support but would need to very closely monitor cost-efficiency, given that even pure training programs spend about two-thirds more per beneficiary than other types of project at the average. 7. Projects should consider setting more ambitious targets for youth support, but recognize that with limited resources, higher youth participation may mean less support to household heads. Jobs projects reach many Nigeriens, with at least 2.6m primary beneficiaries and 11m primary and secondary beneficiaries. However, the level of support to youth is less broad compared to needs, both because very large cohorts of some 500,000 young workers enter the labor market every year (some 7.5% of whom may receive direct jobs support), and because much jobs support is targeted to household heads with the goal of reducing poverty in mind. Policy makers should consider whether it is worth setting more ambitious targets for youth support, recognizing that with limited resources, setting such goals may mean less support to household heads. Household-based programs can inform this debate by studying to what degree youth are able to participate in their activities and to benefit from results achieved. (3) Learning from experience in job support: 8. Projects must do better in collecting information on impacts, sustainability, and cost efficiency of jobs support; this can be done without imposing undue cost. To allow for any real assessment of their performance, projects must track more meaningful indicators of job creation, quality and access than most currently do. To assess cost efficiency, they must keep track of how budget elements link to groups of beneficiaries and to results. Monitoring and evaluation should aim to draw at least low-cost, meaningful comparisons to control groups, for instance by collecting data in a few matched control locations, or comparing to existing data. Further, because the success of jobs projects in Niger depends heavily on whether benefits are sustained, projects should consider simple follow-up monitoring of some key indicators for some beneficiaries; for instance, by phone. In particular, some of the long-standing rural development projects should revisit their earlier efforts and seek to understand how their beneficiaries have fared. 9. Training programs must subject their impacts and cost-efficiency to much more serious scrutiny. While the portfolio of training projects thoughtfully targets market demand, projects provide nearly no meaningful evidence of real jobs outcomes. Given the mixed record of training in promoting such outcomes, better evidence is needed. Projects with very high spending per beneficiary, in particular, must better track outcomes to assess whether the cost incurred is justified. 10. Projects should more explicitly address the gender gap in jobs and incomes and reflect these considerations in project design and outcomes. Most projects reviewed acknowledge the importance of providing better opportunities for women. However, few reflect it in project design, and only some track whether they succeed in supporting jobs for women. A more thoughtful effort is needed to address needs. 11. Projects should systematically track how many women and youth receive support, and what their outcomes are. Many projects show an intention to support youth and women. However, 51 NON-EDITED VERSION to make it possible to assess their performance on these goals, projects must systematically disaggregate their tracking data by age and gender. There is currently a surprising lack of data on how many young Nigeriens and women participate in project activities, and how their results may differ from results obtained among men and older adults. 52 NON-EDITED VERSION REFERENCES Agence Française de Développement. Lutter contre l’exclusion des jeunes dans la région de Diffa Agence Française de Développement. Note de communication Publique d’Opération : CNE 1181 Projet Jeunesse Diffa Agence Française de Développement. Projet d’Appui à la Formation des Ressources Industrielles. Note de Communication d’Opération Publique. Agence Française de Développement. PASAM 2- Appui à la Sécurité Alimentaire au Niger : CNE 1138. Note de Communication Publique d’Opération. Agence Française de Développement (2016). Projet RESILAC (Redressement Economique du Lac Tchad). Note de Prestation d’Opération Agence Nationale pour la Promotion de L’Emploi (2010) Rapport Annuel Agence Nationale pour la Promotion de L’Emploi (2011) Rapport Annuel Aterido, R. and J. Buba (2018). “Firm-level interventions for jobs.” World Bank WP. Bachir, M. (2009). Projet Appui à la Promotion de l’Emploi des Jeunes. Bertrand, M. et al. (2016). “Résultats de l’évaluation d’impact de la composante travaux haute intensité de main d’œuvre du projet Emploi Jeunes et Développement des Compétences en Côte d’Ivoire.” World Bank WP 106574. Bourdain, Y. and Akplogan, B (2010). Evaluation Finale du Programme d’Appui à la Formation Professionnelle et Continue et a l’Apprentissage au Niger. Union Européenne Blattman, C. and L. Ralston (2016). “Generating employment in poor and fragile states: evidence from labor market and entrepreneurship programs.” Boubakar, M. et J. Mivelaz (2012). « Projet Formation et Emploi des Jeunes : Evaluation Externe du Programme ». Swisscontact Boureima, M. Les politiques agricoles au Niger, 1960-2015 : 55 ans à la recherche de la sécurité alimentaire et de la réduction de la pauvreté, les éditions Belle Afrique, 35 pages. Campos, P. (2008). “Evaluation Interne Assistée du programme Infrastructures en Haute Intensité Main d’œuvre (HIMO) dans la zone Téra-Nord » Coopération Suisse Card, D., J. Kluve, and A. Weber (2015). “What works? A meta analysis of recent active labor market program evaluations.” NBER Working Paper 21431. Coopération Belge (2012) Projet de Développement de l’Elevage dans la Region du Liptako Gourma. Rapport d’Achèvement Coopération Belge (2014) Rapport de Résultats 2014 : PAAPSSP Programme d’Appui a l’Aménagement Pastoral et à la Sécurisation des Systèmes Pastoraux. Coopération Suisse (2008). Evaluation Interne Assistée du Programme Infrastructures en Haute Intensité de Main d’œuvre (HIMO) dans la zone Tera-Nord Coopération Suisse (2011). Evaluation Externe Prospective de la Phase 2 du Programme ‘Appui aux organisations rurales pour une sécurité alimentaire effective’. Rapport Final 53 NON-EDITED VERSION Coopération Suisse. Auto-Evalution Mi-Parcours du Programme d’Appui aux organisations Paysannes Phase 3 Coopération Suisse (2016). Auto Evaluation / Capitalisation Mouvement Paysan Nigerien (2013-2016). Coopération Suisse (2017). Rapport final de mise en œuvre du programme d’appui à la petite irrigation (PAPI) DANIDA (2017). Programme de Promotion de l’Emploi et de la Croissance Economique dans l’Agriculture (PECEA). Aide Mémoire de la Revue Mi-Parcours EuropAid OMD1-Pauvreté Article : Protection sociale et emploi – NIGETECH Appui à la formation professionnelle continue au Niger. FIDA (2012a) Projet d’Appui à la Sécurité Alimentaire et au développement dans la région de Maradi (PASADEM). Document de conception de projet FIDA (2012b) Projet de Petite Irrigation Ruwanmu. Document de Conception du Projet FIDA (2013a) Enquête sur l’Impact du Projet de Promotion de L’Initiative Locale pour le Développement a Aigue. Rapport Définitif FIDA (2013b) Projet de Promotion de l’Initiative Locale pour le Développement à Aigue. Rapport d’achèvement de projet FIDA (2014a) Projet d’Urgence pour l’Appui à la Sécurité Alimentaire et le développement rural (PUSADER). Rapport d’Achèvement de projet FIDA (2014b) Projet d’Appui à la Sécurité Alimentaire et au Développement dans la région de Maradi (PASADEM) et Projet de Petite Irrigation Ruwanmu. Aide Mémoire : Mission de supervision conjointe FIDA (2015) Projet de petite irrigation Ruwanmu. Rapport de revue mis-parcours FIDA (2017a) Projet de petite Irrigation Ruwanmnu. Rapport de Supervision FIDA (2017b) Programme de Développement de l’agriculture familiale (ProDaf). Rapport de Supervision FIDA (2018) Rapport d’étude effets/impacts des activités de traitement des bassins versants et de la petite irrigation dans le cadre du projet petite irrigation PPI/Ruwanmnu dans les régions de Tahoua Maradi et Zinder Fox, L. and U. Kaul (2017). “The evidence is in: how should youth employment programs in low-income countries be designed?” USAID. Institut Africain de Développment Durable (2017). « Secteur informel, emploi des jeunes dans les pays fragiles et en conflit armé: étude de cas sur les pays du Sahel ». Institut National des Statistiques (2015) Enquête Nationale d’évaluation d’Indicateurs Socio- Economiques et Démographiques (ENISED) de Janvry, A., E. Sadoulet, and T. Suri (2016). Field experiments in developing country agriculture. Kluve, J., et al. (2019). Do youth employment programs improve labor market outcomes? A quantitative review. World Development, 114, 237-253. 54 NON-EDITED VERSION LuxDev (2016a). Evaluation Finale NIG/017 NIG/817 Appui au Programme national de Formation Professionnelle et Technique d’Insertion des Jeunes sortants. LuxDev (2016b) Rapport de clôture NIG/817 Appui au Programme national de Formation Professionnelle et Technique d’Insertion des Jeunes sortants LuxDev (2016c) Evaluation Finale (NIG/018) Programme d’Appui au Développement Agricole Durable dans la région de Dosso McKenzie, D. (2017a). “Identifying and spurring high-growth entrepreneurship: Experimental evidence from a business plan competition.” American Economic Review, 107(8), 2278-2307. McKenzie (2017b). “How effective are active labor market policies in developing countries? A critical review of the evidence.” World Bank Research Observer. Njonga, B. (2016). Auto-Evaluation /Capitalisation Mouvement Paysan Nigerien 2013-2016 Swisscontact (2015). Evaluation Mi-Parcour dur Programme de Formation Professionnelle Rurale- Phase II (FOPROR II) Niger. Swisscontact (2017). « Programme d’Appui à la Formation Professionnelle Rurale et à l’Employabilité des Jeunes : Rapport de l’Evaluation Finale » Synthèse du rapport d’activités 2008 : Entreprendre au Niger - Structure d’appui à la création et au développement des Micro et Petites Entreprises Mills, B., Premand, P., Stoeffler, Q. (2016). Poor Households’ Productive Investments of Cash Transfers: Quasi-experimental Evidence from Niger. Policy Research Working Paper 7839 UNFPA (2018). World Population Prospects. Union Européenne. Document D’action du Fond Fiduciaire de l’UE. RESILAC : Redressement Economique et Social Inclusif du Lac Tchad . Ref : T05-EUTF-SAH-REG-05 Union Européenne. Document D’action du Fond Fiduciaire de l’UE. Projet d’Appui aux Filières Agricoles dans les Régions de Tahoua et Agadez. Ref : T05-EUTF-SAH-NE-03 Union Européenne. Document D’action du Fond Fiduciaire de l’UE. Création d’Emplois et d’Opportunités Economiques à travers une gestion Durable de l’Environnement dans les Zones de Transit et départ au Niger. Ref : T05-EUTF-SAH-NE 11 Union Européenne. Document D’action du Fond Fiduciaire de l’UE. Appuyer la Formation et l’insertion Professionnelle des Jeunes Filles et Garçons des Régions d’Agadez et Zinder en vue de Contribuer au Développement Socioéconomique de ces deux Régions. Ref : T05-EUTF-SAH-NE 04 World Bank (2010) Amélioration de l’éducation et développement des compétences pour la croissance économique au Niger World Bank (2012a). Rapport d’achèvement: Programme d’Action Communautaires (PAC-2) World Bank (2012b). Competitiveness and Growth Support. Project Appraisal Document (PAD) World Bank (2013) Rapport Analyse Indicative Economique et Financière des Microprojets Réalisés par le PAC 2. 55 NON-EDITED VERSION World Bank (2011a). Community Action Program for Climate Resilience. Project Appraisal Document (PAD) World Bank (2011b). Safety Net Project. Project Appraisal Document (PAD) World Bank (2013b) The Second Phase APL of the Community Action Program. Implementation Completion and Results Report World Bank (2016). Climate-Smart Agriculture Support Project. Project Appraisal Document (PAD). World Bank (2017a). An Assessment of Employment in Niger. World Bank (2017b) Skills Development for Growth Mid-term Review Paper World Bank (2017c). Rapport Interne d’Achèvement : Projet de développement des exportations et des marches Agro-Sylvo-Pastoraux (PRODEX) World Bank (2017d). Rapport Interne d’achèvement du PAC3 du Programme d’Actions Communautaires World Bank (2017e ) . Projet d’appui à la Compétitivité et à la Croissance – PRACC. Mission d’appui à l’évaluation à Mi-Parcours : Rapport de l’Evaluation Interne. World Bank (2018a). Integrated Youth Employment Programs. World Bank (2018b). Accelerating Poverty Reduction in Africa. World Bank (2018c). Niger Community Action Project for Climate Resilience. Implementation Status Report World Bank (2018d). Youth Employment and Productive Inclusion Project. Project Appraisal Document (PAD) World Bank (2019). Doing Business. 56 NON-EDITED VERSION Annex Table 5 Reported short-term increase in revenue among project beneficiaries Number of households Reported relative whose revenues increase Project increased Community Action Program 369,852 80% on average Phase 3 (World Bank) Community Action Program 358,800 At least 30% Phase 2 (World Bank) Ruwanmu Small-Scale 29,700 50% on average Irrigation Project (FIDA) Marketing/Storage/Processing: 72,219 At least 30% Family Farming Development Programme in Maradi, Tahoua and Zinder Regions (FIDA) TOTAL 830,571 Source: World Bank (2012a); World Bank (2017d); FIDA (2018); FIDA (2017b). 57